6 Radio Network Planning
Completion requirements
6 Radio Network Planning
By analyzing radio network planning in the hexagon model, a first simplified overview of principles and measures for capacity improvement can be gained. For real radio networks this model is not good enough. In real networks, cells are not equally distributed and have no regular shape. The assumption of exponential attenuation also is not exact and attenuation is not constant over the whole service area. Radiation is highly anisotropic. Sites are often results of compromises, because not every site can be acquired. Also the telephone and data traffic is inhomogeneous as well as population density.
Thus, for real networks fix reuse patterns are not applicable. In practice, tools are developed based on algorithms and better fitted to the radio network planning needs. The tools are fitted to the network conditions by using measurements.
Radio network planning is kind of a puzzle without the knowledge of the picture the puzzle shows with the constraints that users keep some pieces of the puzzle hidden. Planers need to reshape some pieces if they do not fit exactly.
The hardest part of the planning is gathering all information needed to build a cost effective network with high quality. Unfortunately, some of the information is private and one has to guess the data: Which service mix will be needed, how will users behave? In a surrounding where operators need designs and redesigns very fast there is much unknown data. Many information from different sources have to be considered in the initial network plan on which the rollout depends.These data are needed in different departments, e.g. radio network planning, fixed network planning, realization, controlling.
From the gathered data one can guess the number of base stations needed for a given percentage of coverage of the service area assuming a given load. Experienced planners can derive a first quick and dirty estimate from only a few parameters according to the maximum from the following considerations:
Tab. 6-1: Base station planning in UMTS [UMTSWorld.com].
Thus, for real networks fix reuse patterns are not applicable. In practice, tools are developed based on algorithms and better fitted to the radio network planning needs. The tools are fitted to the network conditions by using measurements.
Radio network planning is kind of a puzzle without the knowledge of the picture the puzzle shows with the constraints that users keep some pieces of the puzzle hidden. Planers need to reshape some pieces if they do not fit exactly.
The hardest part of the planning is gathering all information needed to build a cost effective network with high quality. Unfortunately, some of the information is private and one has to guess the data: Which service mix will be needed, how will users behave? In a surrounding where operators need designs and redesigns very fast there is much unknown data. Many information from different sources have to be considered in the initial network plan on which the rollout depends.
- The business plan of the operator defines, which services the operator wants to offer, how these services shall be implemented and how expensive the roll-out shall be in total.
- The technical realization of the business plan contains service area, grade of coverage, capacity, quality, network features, service mix and a plan for user acquisition.
- The license agreement contains service area and grade, capacity and service evolution, as well as agreements about planned employment numbers and public services as emergency calls.
- The finance plan describes the timing of the rollout.
- The risk analysis shows bottlenecks and critical paths in the project timetable. Mostly this includes site search as the starting point for the further planning. Another important roll plays the transmission capacity over the air interface and the availability of fixed lines to the backbone network.
- Studies about usage of mobile networks underlay market forecast. They include user profiles and geographical user density as well as information about peak hour and peak traffic load.
- From data of the ’statistisches Bundesamt’ the number of inhabitants, income structure, health etc. can be taken. These allow assumptions about the part of the income that can and will be invested into mobile communication.
From the gathered data one can guess the number of base stations needed for a given percentage of coverage of the service area assuming a given load. Experienced planners can derive a first quick and dirty estimate from only a few parameters according to the maximum from the following considerations:
- Capacity needs resulting from planned number of users and service mix in different areas of the service area (BTS capacity planning).
- Grade of coverage resulting from link budgets of high rate data connections.
Because of quality assurance one adds 10% to the sum of sites. Another 10% is added, because not every site can be realized where they are planned. 30% of the final number in general will be cut for cost reasons.
The size of a cell can be determined by a link budget introduced in Chapter 3. The overlap area usually is neglected. Therefore, the number of cells calculated that way has to be increased, because the overlap should take 20-30% of the cell area. Based on the link budget in UMTS in Chapter 3 the base station planning in Tab. 6-1 results. The planning very much depends on the propagation model and overlapping percentage chosen.
| Voice 12.2 k |
Voice 144 k |
|
|---|---|---|
| Cell size from Okumura-Hata model L = 137.4 + 35.2lg(r) |
1.59 km |
0.94 km |
| Covered area of the haxagon |
6.5 km² |
2.3 km² |
| Necessary overlapping of service areas for handover |
20% | 20% |
| Size of the whole service area |
200 km² |
200 km² |
| Number of sites needed |
38 | 105 |
The aim of network planning not only is an initial rollout plan but also its refinement during the lifetime of the system The first plan is only based on estimates, as the user’s behavior occurs only as soon as the network is in use.