3G and 4G are more complex. There are more parameters that you can change in order to optimize network (especially in core network, but also on RAN part of network). For example 2G is using just GMSK-TDMA/FDMA, while in 3G you can use different types of QAM and CDMA. Again in 4G there is OFDM. Those MSC are more complex, but in same time there are many more parameters that you can change. So in 2G you could not change MSC, you could play a little bit with time slots, will it be  speech or data, and that is. Every vendor has it own specific set of parameters with should be in accordance to standards. Based on my experience (while I was working as RAN engineer, we had NSN and Ericsson equipment) sometimes those parameters are not the same, also some algorithms are different. 

As stated previously by Tarik, network/cell parameters are more complex (at least in 3G; 4G is a whole new case, as functionalities such as SON can decrease optimization and planning tasks to some extent). As a really basic comment, I would say:

2G: you provide frequency plans for BCCH and TCH channels. You also work on intra-RAT HO.

3G: you provide Scrambling Code planning, CPICH power planning distrbution. In terms of HO, you now work with SHO, as well as with Inter-RAT HO.

This is really very simple and only on the radio side. Hope it can help. Good luck!

All the best,

Francisco

@ Francisco Falcone, @Tarik Kazaz  thanks for your answers. Since I am Computer science student you are talking too deep notions in RF. What I ment is that during the coverage planning of 2G and 3G, 4G are the parameters (of the antenna) that need to be optimised the same ?.

Best Regards

Hi again. When you ask for antenna parameters, do you mean gain, horizontal and vertical beamwidth, sidelobes and front to back lobe level? If this is the case, it really depends on the coverage area under analysis. The main difference between 2G adnd the rest is the fact that 2G works under absolute receiver sensitivity levels, whereas 3G and beyond work under coverage/capacity relations and hence, are stronlgy dependent on interference levels. A very simple and coarse estimation:

- For urban macrocell scenarios: sector antennas, with gains in the order of 14-17dBi, horizontal beamwidth between 60º and 90º, vertical beamwidth in the order of 5º-7º, front to back lobe levels in the order of 20-25dB

-Microcells: usually small panels or bi-lobular antennas

- Picocells/Femtocells: short vertical monopoles

Of course, in the BTS/NodeB/eNodeB you might have RX diversity. MIMO is really not an option nowadays (in any case, SIMO).

Again, hope it can help!

@Francisco Falcone , Thanks for your answer. Do you think that a mathematical model (formulation) can be proposed to be applied for covergae optimisation in 2G, 3G and 4G also ?

If yes, references of papers doing this will be welcome !! 

If no, references of papers proposing models for coverage optimisation in each generation will be welcome also. 

Thanks in advance.

Well, there are some references, related with Radio Network Optimization. I will send them ASAP.

All the best,

Francisco

Zakaria A. D. for planning in 2G (BCCH/TCH) in 3G by PSC in 4G by PCI