CDMA timing Odds and Ends.
Scanned from a text book that seems reliable. Equations may not be correct, since they do not scan well.
snip
In cdmaOne and cdma2000, the mobile uses the following three search windows to track the received pilot signals:
SRCH- WIN-A: search window size for the active and candidate sets
SRCH- WIN-N: search window size for the neighbor set
SRCH- WIN-R: search window size for remaining set
SRCH-WIN-A
SRCH- WIN-A is the search window that the mobile uses to track the active and candidate set pilots. This window should be set according to the propagation environment. This window should capture all usable multipath signal components of a base station, and, at the same time, the window should be small enough to maximize searcher performance.
The search window should be large enough to accommodate the maximum expected arrival time difference between the pilot's usable multipath components (i.e., the pilot's maximum delay spread).
T d,A > 2 x (td) max /Tchip
where
Td, A = required delay budget (chips)
(td)max = maximum delay spread in seconds
T chip = chip time (813.8 nanoseconds)
EXAMPLE
Consider the propagation environment of a cdma2000 network, where the signal with a direct path travels 3 km to the mobile, and the multipath component travels 5 km before reaching the mobile. What should be the size of SRCH- WIN-A ?
Direct path travels a distance of = 3000/244 = 12.3 chips
Multipath travels a distance of = 5000/244 = 20.5 chips
The difference in distance traveled between the two paths 20.5- 12.3 = 8.2 chips
The window size ~ 2 x 8.2 = 16.4 chips
Use window size = 17 chips
EXAMPLE
Consider cells A and B of a cdma2000 network, separated by a distance of 20 km. The mobile travels from cell A to cell B. The RF engineer wishes to contain the soft handoff region between points X and y located at distances 8 and 12 km from cell A.
What should be the search window size?
At point X the mobile is 8,000/244 = 32.8 chips from cell B
At point X the mobile is 12,000/244 = 49.2 chips from cell A
Path difference = 49.2- 32.8 = 16.4 chips
At point y the mobile is 12,000/244 = 49.2 chips from cell B
At point y the mobile is 8,000/244 = 32.8 chips from cell A
Path difference = 49.2- 32.8 = 16.4 chips
The SRCH- WlN-A > 2 x 16.4 > 32.8 chips (use window size = 33 chips)
SRCH-WIN-N and SRCH-WIN-R
SRCH- WIN-N is the search window that is used by the mobile to monitor the neighbor set pilots. The size of this window should typically be larger than that of SRCH- WIN-A. SRCH- WIN-R is the search window that the mobile uses to track the remaining set pilots.
A typical requirement for the size of this window is that it should be at least as large as SRCH-WIN-N.
The SRCH- WIN-N and SRCH- WIN-R pilots should account for the largest delay spread of the target pilot as well as the largest difference in propagation delays (i.e., difference in distance) between the reference pilot and target pilot.
T d N = T d R > (2xD max /Vc+(t)max )/ Tchip
where
Dmax = maximum difference (in miles) between (1) the mobile and cell transmitting active set pilot and (2) the mobile and the cell transmitting neighbor (or remaining) set pilot
Vc = speed of light (186,000 miles per second)
EXAMPLE
A CDMA system is located in an urban area where the maximum measured delay spread is S ~s. The maximum distance between two neighboring base stations is 1.2 miles. Determine the active and remaining set window sizes.
TdA>2X(td)max/Tchip = 2x5x10e-6/813.8 X 10e-9 = 12.29~13 chips
:.(SRCH-WIN-A) = 13 chips
Td,N = Td,R > (2 X Dmax/Vc + ('td)max)/Tchip = (2 X 1.2/186000 + 5 X 10e-6 )/ 813.8 X 10e-9 = 22 chips
:.(SRCH-WIN-R) = 22 chips
CDMA system planning and optimization has many more degrees of freedom than AMPS or TDMA-based systems. Additional degrees of freedom are a blessing and a curse. Some CDMA parameter adjustments allow manipulation of cell coverage areas to cover flaws in the initial design without physically moving cells. On the other hand, traffic could create coverage gaps in city centers during peak usage periods if the system is improperly configured. A full understanding of the properties of CDMA technology is vital to the successful design and operation of CDMA networks.
Proper PN offset planning is critical to avoid alias PN sequences in a IS-95/cdma2000 networks. The PN offset planning is relatively simple for a network in which most of the cells are of same size. In a typical network, the cells in rural and highway areas are much larger compared with the cells in urban areas. For such a network, PN offset planning becomes important. The Co-PN offset alias is the major concern for the small cell cluster, whereas adjacent PN offset alias is more likely to occur in large cells. There are three types of PN offset confusion problems: Co-PN, adjacent PN, and confused handoff. PN confusion is more harmful than conventional interference.
.Propagation delay, search window sizes, and path-loss exponent are three key factors in
PN offset planning. Sufficient reuse distance and PN separation, proper antenna downtilt, and proper allocation of PN offset values reduces the possibility of PN confusion.
For nonuniform networks, reuse pattern clusters for large cells and small cells may have different shapes and may use PN offsets from two disjoint sets; small cell clusters use PN offsets with smaller values and large cell clusters use larger offset values. Other significant issues for CDMA systems are soft handoff optimization, reverse channel noise floor management, the impact of geographical user distribution within a cell, and the criticality of power management.
In this chapter, coverage of a loaded CDMA network was studied by analyzing the uplink range. Base station multiuser detection (MUD) receiver can provide good coverage even with high system load after initial deployment. MUD decreases mobile station transmission power in a loaded network. Therefore, a network with base station MUD can be operated with a higher percentage of maximum load if the system capacity is limited by downlink. The coverage need not be sacrificed in order to use high system loads. In CDMA network planning, the performance of MUD should be considered when predicting and planning the coverage. The effect of MUD on cell range also depends on the propagation environment. Higher-data-rate services will reduce cell range in the uplink as the transmission power is limited. Therefore, in the cell design, the coverage area for low-rate services will be different than for high-rate services.
snap
WCDMA had both an async and sync mode. Certain groups do not trust GPS, that evil American system, and do not want to use it for sync. However, accurate timing has been around for decades before GPS. Datum (http://www.datum.com ) has produced equipment for years and many years ago the timing was quoted as accurate within 9 nanoseconds of the correct time. Several times a year, a flying clock from the National Bureau of Standards would be used to calibrate the local clock. One form of the time available is IRIG B. (http://www.telser.net/products.htm , use Google to find many more). I see that even TDMA base stations use GPS to keep the oscillators accurate.
Async WCDMA has to use much wider search windows to find the other base stations. I do not have the numbers handy now but remember seeing some rather large numbers which require more handset compute time. |
