Optimization of 2G/3G interoperability parameters in WCDMA deep coverage scenarios

Welcome to the IT technology community forum, and interact with 2 million technical staff. Table 1 Comparison of WCDMA and GSM signal distribution from the relationship between RSCP and EcIo, when RSCP approaches-100dBm, the corresponding EcIo is-7 ~ -9 dB, and EcIo will reach below-14dB when RSCP is below-110dBm; when the pilot signal is clean, RSCP is-105dB

Welcome to the IT Community Forum and interact with 2 million technical staff> Table 1 Comparison of WCDMA and GSM signal distribution from the relationship between RSCP and Ec/Io, when RSCP approaches-100 dBm, the corresponding Ec/Io is-7 ~ -9 dB, and when RSCP is below-110 dBm, Ec/Io will reach below-14 dB; when the pilot signal is clean, RSCP is-105 dB

Welcome to the IT Community Forum and interact with 2 million technicians>

Table 1 Comparison of WCDMA and GSM Signal Distribution



From the relationship between RSCP and Ec/Io

-The Ec/Io corresponding to 100 dBm is-7 ~ -9 dB. When RSCP is below-110 dBm, Ec/Io will reach below-14 dB. When the pilot signal is clean and RSCP is below-105 dBm, ec/Io will reach about-9 dB.
The signal strength of RSCP in most areas (including study, guest bedroom, living room, master bedroom, and kitchen) is between-95 and ~ -Between 105 dBm, the signal between the aisle, the customer and the storage is slightly different, while the Ec/Io in most regions is above-14 dB, in most indoor areas, the GSM signal strength is below-80 dBm.



A) reselect the policy. When the Ec/Io of the service community
B) Switching policy. When Ec/Io is measured
According to the previous signal distribution analysis, in this scenario, the GSM signal is much better than the WCDMA signal, so as long as the different system search or measurement is started, reselect or switch (this is also the reason why the 3A event is not listed or related parameters are reported periodically ). However, the threshold for enabling reselect search is Ec/Io <-14 dB, and the threshold for enabling compression mode switching measurement is RSCP ≤-101 dBm or Ec/Io ≤-15 dB. According to the relationship between RSCP and Ec/Io, it is not difficult to see that the signal is-101 ~ -105 dBm, the Ec/Io value is greater than-10 dB, so the phone will not reselect in such areas, and once the call starts, it will immediately reach the RSCP threshold of the 2D event, switch will occur. In addition, RSCP is-95 ~ -The signal of 100 dBm is a weak signal, and the signal itself fluctuates greatly. Once weak, it switches to 2 GB. Currently, CS switching is one-way, so the end of the call is at 2 GB. Table 4 shows current network parameter tests.



Table 4 Current Network Parameter Tests

Table 4 shows that the receiving level is-100 ~ -108 dBm, the Ec/Io values do not reach the-14 dB threshold. These calls are switched immediately after the start, and the receiving level is-93 ~ due to floating signals ~ -The signal of 100 dBm will also go through several seconds after the original position is not moved.

2. Experiment verification of optimization parameters

According to the results of the current network parameter test, when the current network parameter is configured, the user will immediately switch to GSM after the call is initiated in most areas such as indoor, living room, and study rooms, the test results conform to the theoretical analysis.

There are two ways to solve the problem.

A) Let the terminal reselect to 2G network earlier.

B) enable the terminal to initiate WCDMA → GSM switchover later.

In the first method, the RSCP corresponding to Ec/Io must be higher than-100 dBm. From the signal quality distribution, the SsearchRAT Threshold needs to be raised to-5 ~ -6 dB, which is obviously unrealistic. This will cause the user to re-select the GSM network too early, greatly reducing the coverage of the WCDMA idle mode. The best way is to reduce the threshold of measurement startup or disable RSCP to trigger the compression mode, and reduce the probability of switching between different systems. After modifying the parameters, a parameter test is conducted to adjust the threshold of 2D events that start the compression mode.

2.1 test parameter 1 Test Result

Parameter 1 mainly changes the cpich rscp 2D threshold from-100 dBm to-108 dBm, so that the indoor signal sampling can basically be guaranteed to switch immediately after the call is not started, if the signal is lower than-109 dBm, the Ec/Io will almost reach-14 dB, and the 2G network will be reselected before the call, parameter 1 mainly describes the BLER before the switch and the success rate of the switch. In order to ensure the adaptability of parameters, when the signal enters the corner or there is an object occlusion, the signal is quickly weakened during the moving switching process ~ Whether to drop the call in the case of 20 dB. The test result is shown in table 5.



Table 5 test results of test parameter 1

As can be seen from table 6, after the parameter is modified, most of the points do not have the problem of switching immediately after the start of the call, but need to move to the weaker coverage area to switch, some signal points below-105 dBm will be switched due to signal fluctuation, and the-109 dBm start pressure mode will be modified, and the switching will not be triggered due to rapid signal fading, the BLER value before the switch is basically 0, indicating that the speech quality is good, so this threshold is the acceptable threshold. It can be seen that most of the indoor areas do not switch to GSM immediately after the parameter is modified.

2.2 parameter 2 Test Results

Compared with parameter 1, parameter 2 only modifies the Trigger Time delay of 2D events To Trigger 2D, mainly To test whether parameter 2 increases the probability of disconnection. The sensitivity of parameter 2 was tested 11 times in different locations, of which four were dropped.

It can be seen that after parameter 2 is modified, the dropped call is obvious, and the dropped call rate reaches 30%. After parameter 2 is modified, after RSCP reaches-109 dBm, it still needs to wait 1 280 ms to start the measurement of the different system. In this process, the probability of disconnection is increased due to the further deterioration of the mobile signal. Test Description: After the RSCP threshold of 2D is reduced, the margin left by itself is small. If the delay trigger time is too long, the signal will deteriorate and the switchover will fail, in the high-speed mobile scenario, you should reduce the parameter value and complete the measurement and switching earlier.

3 conclusion

Based on the above test analysis, we can draw the following conclusions.

A) In the current network coverage, outdoor signals are generally distributed between-75 and ~ -Signal Strength in residential buildings is distributed between 85 dBm

-95 ~ -105 dBm, which is also the main RSCP distribution range in the current indoor depth coverage area. According to the relationship between RSCP and Ec/Io, when the network load is low, the corresponding Ec/Io is about-7 ~ When RSCP approaches-100 dBm ~ -9 dB, and when RSCP is below-110 dBm, Ec/Io will reach below-14 dB. When RSCP is-105 dBm, Ec/Io reaches-7 dB without pilot pollution. The deterioration rate of Ec/Io is much lower than that of RSCP.

B) in the case of total station coverage, the loss of 3G indoor signals is much greater than 2G. The 3G signal in the indoor depth coverage area is weak, but the GSM signal remains good. When the 3G signal is lower than-100 dBm, the GSM level is still near-70 dBm. When the 3G signal reaches-110 dBm, the GSM level is still between-80 and ~ -Between 85 dBm.

C) At present, there is still margin for switching the CS Service's current network threshold. There is a problem with the combination of reselect parameters and voice switching parameters, and related parameter optimization is required. Considering that WCDMA has strong support for voice services, there is still a margin for lowering the threshold for measuring the startup of different systems on the current network. Considering the impact of load increase in the future, we recommend that the startup threshold of different systems (generally 2D) be controlled at-105 dBm. The Time To Trigger of different system delays should not be too low. It is recommended that the value not exceed 640 ms in Urban Deep coverage scenarios.

4 Conclusion

Although this article only studies a deep coverage scenario that has not been divided into rooms as an example, it provides parameter optimization suggestions for this scenario. However, considering the coverage of China Unicom's network, it also has some promotion significance. It should be particularly noted that the parameter modification of a cell should take into account multiple factors. The cell coverage of the same scenario may include multiple scenarios with different parameter requirements, therefore, the optimization of interoperability parameters should also be fine-tuned based on the actual situation. While ensuring the 2G/3G interoperability performance, the WCDMA network should be maximized, 3G users should try their best to use 3G network services.

[1] [2]