There are three bit error measurements available in the test set:
There are four commonly used types of bit error measurements in GSM:
The test set allows you to select between loopback A, B or D for Bit Error Measurement. Loopback C can only be used in the fast bit error measurement. Refer also to Fast Bit Error Measurement Description .
| NOTE |
Previous versions of the test set required you to set the loopback type, and did not have a feature that allowed time for the loop to close.
During BER measurements, the test set generates a downlink TCH with pseudo-random binary sequence, PRBS-15, data at a known power level. The mobile station receives the data, loops it back to its transmitter, and returns the data to the test set. The test set compares data sent to data received, and BER is calculated.
SETup subsystem commands are sent to the test set to specify the time taken to close its loopback path, whether to open or close a loop during downlink signaling operations (for example, channel assignment), the number of bits to test, measurement type, loopback delay, measurement units, trigger arm, and measurement timeout values.
When a call is established on a TCH, the loopback type, corresponding to one of the BER measurement types, must be sent to the mobile station. The test set closes the loopback automatically and re-opens it when the measurement is closed (that is, when INITiate:BERRor is OFF).
You must set the measurement type from one of the available measurement types, (see SETup:BERRor[:TYPE] ). If you query a residual result when a non-residual measurement is initiated, the test set returns 9.91 E+37 (NAN). The Measurement type must be set before initiating a BER measurement. See BER Measurement Types
The loop must be closed before a BER test can start. Using the close loop signalling delay time feature allows time for the loop to close. See SETup:BERRor:CLSDelay[:STIMe] for more details.
Each mobile station may have a different time delay between receiving a frame and re-sending it on the uplink. By default, the test set is configured to automatically determine the amount of delay needed. This is determined automatically when the test set has, for two frames, correctly received 80% of the downlink bits back on the uplink. The test set can be queried for the loopback delay value.
If necessary, you can manually set the loopback delay (see SETup:BERRor:LDControl:AUTO ).
| NOTE |
The BER measurement trigger source is always set to immediate. The BER measurement does not offer multi-measurement results. See Statistical Measurement Results
BER, FBER, and DAUDIO (uplink speech level) measurements are mutually exclusive measurements. Whichever of these measurements is activated last forces the others to stop.
Type IA, Type IB, and Type II for both residual and non-residual refer to the different class types describing GSM speech coding bits.
When one of these residual measurement types is selected, the test set sends the mobile station a message instructing it to close loopback A. When the mobile station is in loopback A, it erases frames received with CRC (cyclic redundancy check) errors and replaces all the bits in those frames with zeros. The BER measurement is then able to return the frame erasure count or ratio results as well as the results of the bitwise comparison of the un-erased frames looped back from the mobile station.
When you select one of these non-residual measurement types, the test set sends the mobile station a message instructing it to close loopback B. A BER measurement with CRC's (cyclic redundancy check) returns the CRC count or ratio results. The mobile station does not indicate if any frames in the downlink were erased.
When one of these residual (D) measurement types is selected, the test set sends the mobile station a message instructing it to close loopback D. The test set works almost the same as the correspondent Residual measurement type with one exception: the FER display during the measurement is actually Frame Error Rate plus Unreliable Frames Rate.
The results of a BER measurement can be displayed in two ways, (number of errors counted) or (the ratio bad bits (errors) to total bits counted). If you are using the test set manually, select either Count or % from the Measurement Units field. If you are using the test set remotely, these results are available using the FETCh command, see FETCh:BERRor:COUNt[:BITS]? or FETCh:BERRor:RATio[:BITS]? . Alternatively the FETCh:BERRor[:ALL]? or FETCh:BERRor:FULL? can also be used to return the results.
The GSM bit error measurement is also useful for CSD bit error testing. For more information on GSM CSD, see GSM Circuit Switched Data (CSD) .
You must start a CSD call before initiating the GSM BER measurement.
Since all the data in the CSD data payload is equally important and convolutionally encoded with same protection, the test set allows you to choose two new measurement types to ensure appropriate results are returned: Residual Data and Data.
In addition, the frame-by-frame Fast Bit Error measurement is also applicable to CSD bit error testing.
GSM Bit Error Measurement is available when the operating mode is set to Active Cell, GSM BCH+TCH, GPRS BCH+PDTCH+TCH and EGPRS BCH+TCH+PDTCH.
Programming a Bit Error Measurement
Fast Bit Error Measurement Description
During GBER measurements, the test set generates one or more downlink PDTCHs at a known level with data using the payload pattern you specify. (See Payload Patterns .) The default payload pattern is a pseudo-random binary sequence, PRBS-15, which is the most appropriate for this measurement. The mobile station receives the data, loops it back to its transmitter, and returns the data to the test set. The test set compares data sent to data received, and BER is calculated.
If the test set's operating mode is set to Active Cell, the connection type must be set to ETSI Type B (Unacknowledged) or ETSI Type B (Acknowledged) in order to make BER measurements. This can be done using CALL:FUNCtion:CONNection:TYPE . The measurement does not return any results if the connection type is set to SRB Loopback.
Alternatively, you can make BER measurements when the test set's operating mode is set to GPRS BCH+PDTCH or EGPRS BCH+PDTCH test modes (see CALL:OPERating:MODE ).
SETup subsystem commands are sent to the test set to specify the loopback delay control mode, number of bits to test, block delay, measurements units, trigger arm, measurement timeout values, and bad block handling setting.
The bad block handling parameter determines how bad blocks are interpreted by the GBER measurement. If you want more details on this parameter, see SETup:GBERror:BBLocks .
Each mobile station may have a different time delay between receiving a block of data and re-sending it on the uplink. By default, the loopback delay control mode is set to automatic, and the amount of delay needed is determined when the test set has, for four consecutive blocks, correctly received 80% of the downlink bits back on the uplink. The test set can be queried for the block delay value using FETCh:GBERror:DELay? .
You can specify the first downlink burst to be looped back in the first uplink burst using the command CALL:(PDTCH|PDTChannel):MSLot[:FIRSt]:DOWNlink:LOOPback[:BURSt] .
If necessary, you can manually set the delay using SETup:GBERror:LDControl:AUTO .
| NOTE |
The GBER measurement trigger source is always set to immediate. The GBER measurement does not offer multi-measurement results. If you require more details on multi-measurement results, see Statistical Measurement Results .
The GBER measurement aborts and restarts if the Operating Mode, Multislot Configuration, or Coding Scheme is changed while the GBER measurement is running.
Any change to the Multislot Configuration or the Coding Scheme affects the rate at which bits are tested. More uplink timeslots per frame will increase the number of bits tested per second as well as coding scheme used. For example, if the coding scheme CS-4 is selected on the test set, the GBER measurement can test more bits per second than if the coding scheme CS-1 is selected. Therefore, the recommended channel coding scheme for GPRS BER measurements is CS-4. Use the command CALL:(PDTCH|PDTChannel):CSCHeme:UPLink to set the GPRS coding scheme.
The GBER and the BLER measurements are mutually exclusive, that is which ever of these measurements is activated last forces the other to become inactive.
GSM Bit Error Measurement is available when the operating mode is set to Active Cell, GSM BCH+TCH, GPRS BCH+PDTCH+TCH and EGPRS BCH+TCH+PDTCH.
GBER measurements can be made on these types of input signals:
The results of a GBER measurement can be displayed in two ways; the number of errors counted or the ratio of bad bits (errors) to total bits counted. If you are using the test set manually, both Count and Bit Error Ratio are displayed on the Bit Error measurement screen. If you are using the test set remotely, these results are available using the FETCh command, see FETCh:GBERror:COUNt? or FETCh:GBERror:RATio? . Alternatively the command FETCh:GBERror[:ALL]? can be used to return all of the following results:
In addition, the Intermediate Count of Bits Tested result can be queried using FETCh:GBERror:ICOunt? , the Block Delay can be queried using FETCh:GBERror:DELay? , and the CRC error count can be queried using FETCh:GBERror:CRC? .
During the SBER measurement, the test set generates one or more downlink PDTCHs at a known level with data using the payload pattern you specify. (See Payload Patterns .) The default payload pattern is a pseudo-random binary sequence, PRBS-15, with is the most appropriate for this measurement. To accurately measure bit error, the measurement must reflect the bit error rate before error correction is applied. For EGPRS, this is provided by implementing the ETSI defined SRB Loopback test mode in the mobile station. In this test mode, the mobile station loops back the received downlink PDTCH after demodulation but before decoding, thus bypassing the error correction stage. The mobile station receives the data, loops it back to its transmitter, and returns the data to the test set. The test set compares data sent to data received, and BER is calculated.
The SBER measurement is made on the raw data after coding and before decoding. This means not all the bits returned by the mobile station are included in the comparison. Only the encrypted bits which are not "stealing bits" are included in the calculation. Therefore only the following bits are used in the SBER measurement:
Asymmetric SRB (ASRB) loopback mode is defined under the circumstance that the MS is unable to support 8PSK modulation capability on the uplink. In this scenario the bits on 8PSK downlink PDTCH burst are looped back over three successive GMSK uplink bursts. The radio block sent on a downlink timeslot using 8PSK modulation should be followed by two radio blocks on the same timeslot where GSM dummy bursts are transmitted, and MS shall not retransmit the last 16 encrypted bits received in an 8PSK radio block when retransmitting it using GMSK modulation.
In order to let MS go to ASRB loopback mode, the corresponding mode flag in the layer 3 message should be set and sent to MS at the beginning of starting a data connection, otherwise MS will not go to ASRB mode by any change. For example, changing downlink coding scheme to 8PSK and uplink to GMSK in transferring will not get the right results.
If the test set's operating mode is set to Active Cell, the serving cell must be set to EGPRS and the data connection type must be set to SRB Loopback in order to make SBER measurements. This can be done using CALL:FUNCtion:CONNection:TYPE .
Alternatively, you can make SBER measurements when the test set's operating mode is set to EGPRS BCH+PDTCH test mode (see CALL:OPERating:MODE ).The SBER measurement will work correctly in this mode if the modulation format used on a downlink slot has the same modulation format as the corresponding uplink slot. If the SRB Loopback mode is supported, the SBER measurement will also function when an 8PSK downlink modulation format is looped into a GMSK uplink modulation format.
For ASRB loopback mode BER measurement, the result could either synchronize incorrectly or return high BER when the 8PSK uplink capable MS is used, As there is no phone can support both SRB and ASRB, do not change the PDTCH coding scheme type from SRB mode to ASRB mode, or opposite, during the data connection, otherwise the SBER result could be incorrect
SETup subsystem commands are sent to the test set to specify the loopback delay control mode, number of bits to test, block delay, measurements units, trigger arm, and measurement timeout values. See SETup:SBERror .
Each mobile station may have a different time delay between receiving a block of data and re-sending it on the uplink. By default, the loopback delay control mode is set to automatic, and the amount of delay needed is determined when the test set has, for four consecutive blocks, correctly received 80% of the downlink bits back on the uplink. The test set can be queried for the block delay value using FETCh:SBERror:DELay? .
You can specify the first downlink burst to be looped back in the first uplink burst using the command CALL:(PDTCH|PDTChannel):MSLot[:FIRSt]:DOWNlink:LOOPback[:BURSt] .
If necessary, you can manually set the delay using SETup:SBERror:MANual:DELay .
| NOTE |
The SBER measurement trigger source is always set to immediate. The SBER measurement does not offer multi-measurement results. If you require more details on multi-measurement results, see Statistical Measurement Results .
| NOTE |
SBER measurements can be made on these types of input signals:
The results of an SBER measurement can be displayed in two ways; the number of errors counted or the ratio of bad bits (errors) to total bits counted. If you are using the test set manually, both Count and Bit Error Ratio are displayed on the EGPRS SRB Bit Error measurement screen. If you are using the test set remotely, these results are available using the FETCh command, see FETCh:SBERror:COUNt? or FETCh:SBERror:RATio? . Alternatively the command FETCh:SBERror[:ALL]? can be used to return all of the following results:
In addition, the Intermediate Count of Bits Tested result can be queried using FETCh:SBERror:ICOunt? and the Block Delay can be queried using FETCh:SBERror:DELay? .
Programming a Bit Error Measurement
How do I make EGPRS SRB Bit Error Measurements?