Frame Error Rate (FER) Measurement Description

Last updated: November 22, 2010

How is an FER measurement made?

Frame Error Rate (FER) measurement is used to test the performance of a mobile station's receiver. FER measurements can only be made on the forward traffic channel (F-Traffic) for IS-95 system or forward fundamental channel (F-FCH) for IS-2000 system and a call must be connected with a loopback service option (such as SO2, SO9, SO55 or SO75). The forward supplemental channel (F-SCH) for IS-2000 system is measured using TDSO FER measurement (see TDSO Frame Error Rate Measurement Description ).

During an FER measurement, the test set sends a sequence of frames to the mobile station. Each frame contains CRC (Cyclic Redundancy Code) bits, which provide frame quality indicator and allow the mobile station to verify that it has correctly decoded a frame.

The mobile station is put into a loopback service option and makes its best attempt to decode each received frame sent from the test set. Once the mobile station determines the CAT_TYPE that specifies whether the frame received is a good frame, bad frame, frame erasure, or a frame blanked by signaling, the mobile station encodes and re-transmits the frame, with the first two bits replaced with the CAT_TYPE information, back to the test set. The test set compares each received frame to the corresponding frame that was sent and validates the CAT_TYPE information, then determines the measurement results (see FER Measurement Results ).

The test set keeps a running count of the measured frames and the number of frames that contain bit errors.

Intermediate measurement results are periodically available on the front panel display, but are not available programmatically through the GPIB interface.

Confidence Level Testing

Confidence level testing is a feature of FER measurements that applies statistical analysis to FER measurements so that pass/fail test results can be obtained in the shortest possible time.

When confidence level is on, a minimum number of frames as specified must be measured before confidence algorithm can begin (For fading tests, the minimum test duration is required to meet the power measurement uncertainty requirement of 0.2 dB). After the measurement is initiated and the specified minimum frames have been tested, given an FER requirement and confidence level settings, a calculation is made to determine the minimum number of remaining frames necessary for a valid statistical analysis. At that time it is also determined what the maximum number of bad frames received would be necessary to fail the test. If the pre-calculated maximum number of bad frames is received before the minimum remaining frames is measured, the FER measurement will ends with a fail indication.

When confidence level is on, the FER measurement will keep track of the number of frames measured, the number of frame errors, and perform a statistical analysis of the probability that the value entered in the FER requirement field will (or will not) be exceeded. The confidence level parameter is typically set at 95%. With that setting, the FER measurement returns a Pass confidence result before the maximum frame count is reached if statistically there is a 95% probability that the percentage of frame errors counted will not exceed the specified FER requirement value if the measurement is allowed to reach the maximum frame count. The FER measurement returns a Fail confidence result before the maximum frame count is reached if statistically there is a 95% probability that the percentage of frame errors counted will exceed the specified FER requirement value if the measurement is allowed to reach the maximum frame count. If the number of frame errors counted does not allow a reliable prediction of Pass or Fail confidence result, the measurement continues until the number of frames tested equals the specified maximum frame count. "Max frames" and the FER percentage results are then returned.

If confidence level is off, the measurement always runs until the number of frames tested equals the specified maximum frame count. "Max frames" and the FER percentage result are then returned.

FER Measurement Specific to the Lab Application

The lab application can perform mobile station tests that require faded signals. This setup requires a rear panel digital bus connection to an external PC running Baseband Studio software. See Digital Bus Information for details.

Whenever the test set is configured to generate faded forward channel signals with AWGN, an external AWGN value is displayed on the FER measurement screen.

When working in forward power control (FPC) mode 000, the mobile station uses the 800 bps (400 bps in 1x advanced) primary reverse power control subchannel to control the F-Traffic/F-FCH power level. The performance of forward power control is measured by the differences between the mobile station received FER and the assigned target FER while not exceeding the required Eb/Nt limit. For more information on how FPC works, see Forward Power Control Overview .

When the FPC mode 000 is enabled on the test set (see FPC Mode ), the mobile station can direct the test set to change F-Traffic/F-FCH level to achieve a certain FER. If the F-Traffic/F-FCH level reaches a very high value, Eb/Nt may exceed the Eb/Nt limit as specified by the test standard but the FER confidence result is Pass. To solve this problem, the test set will calculate the average Eb/Nt, maximum Eb/Nt and minimum Eb/Nt of the F-FCH during the FER measurement and determine the Eb/Nt Pass/Fail result when the measurement completes. If the average Eb/Nt is no greater than the specified Eb/Nt Limit, the Eb/Nt Pass result is returned to indicate that the mobile station has passed the Eb/Nt limit for the FER test. Providing the Eb/Nt Pass/Fail result in addition to the FER confidence Pass/Fail result can help you identify which one leads to the final FER failure. All these results are displayed on the FER measurement screen (see FER Measurement Results Specific to Lab Application ).

NOTE

The initial stage of the FER measurement algorithm is to determine the loopback delay through the mobile station. This synchronization stage requires that a small number of consecutive frames matching those sent by the test set be looped through the mobile station and received at the test set.

In the lab application, when the F-FCH/Traffic Frame Pattern State is set to On and the F-FCH/Traffic Frame Pattern Good Frames is set to a value less than three the FER measurement may not be able to synchronize with the mobile station. An error message will be generated if the F-FCH/Traffic Frame Pattern State is set to On, the F-FCH/Traffic Pattern Good Frames is set to a value less than three, and the FER measurement is running.

FER Measurement Parameters

Confidence State - Controls whether the confidence level will be used during the FER measurement. See Confidence Level Testing for details.
Confidence Level - Sets the confidence level (range is 80% to 99.99%) that is used to reduce the test time for the specified FER Requirement. See Confidence Level Testing for details.
FER Requirement - Sets the percentage of frame errors (range is 0.1% to 30%) that must not be exceeded for a confidence Pass result to be returned. When the Confidence Level is on, the FER Requirement is used to determine how many frames to test before making pass/fail results available. See Confidence Level Testing for details.
Maximum Frame Count - Sets the maximum number of frames (range is 25 to 10,000,000 frames) to be tested during the FER measurement.
Minimum Frame Count - Sets the minimum number of frames (range is 0 to 10,000,000 frames) to be tested before applying the confidence algorithm when the Confidence State is on. It does not apply when Confidence State is off. See Confidence Level Testing for details. Values for minimum frame count are specified in the standard for the faded tests.
First PCG Index - It's available when system type is IS-2000, and radio configuration is (Fwd11, Rvs8).
Second PCG Index - It's available when system type is IS-2000, and radio configuration is (Fwd11, Rvs8).
F-FCH/Traffic Level - See Configuring Forward Channel Power Levels and Walsh Codes .
F-FCH ACK Mask- No RL Blanking, Upper- See F-FCH Ack Mask- No RL Blanking
F-FCH ACK Mask- No RL Blanking, Lower- See F-FCH Ack Mask- No RL Blanking
Forward N2m Indicator- See Forward N2m Indicator
R-FCH Blanking Duty Cycle- See R-FCH Blanking Duty Cycle
Rate 1/8 Non-Critical Frames Ratio- See Rate 1/8 Non-Critical Frame Ratio
Eb/Nt Limit ( for lab application only ) - Sets the Eb/Nt limit (range is -30 to 42 dB) that must not be exceeded for a Eb/Nt Pass result to be returned. This setting applies only when the FPC mode 000 is enabled on the test set. See FER Measurement Specific to the Lab Application .
Trigger Arm - (See Trigger Arm (Single or Continuous) Description ).
Timeout Time/State - (See Timeouts ).

FER Measurement Results

Integrity Indicator - (see Integrity Indicator )
Confidence Pass/Fail - indicates whether the measurement achieved confidence of failure or success, or whether maximum frame count was reached. See Confidence Level Testing .
FER - the ratio of Total Frame Errors to Frames Tested expressed as a percentage.
Total Frame Errors - the total number of frames found in error including Mobile Errors, Forward Erasures and Reverse Erasures.
Forward Erasures - the number of frames that the mobile station could not decode. These frames are reported as CAT_TYPE 11 by the mobile station.
Reverse Erasures - the number of frames that the test set could not decode. This is caused by poor transmitted signal quality from the mobile station.
Mobile Errors - errors detected by the test set via a bit-by-bit comparison of transmitted frames and the frames looped back from the mobile station. These frames can be the mobile reported bad frames (CAT_TYPE 10) which are indeed bad frames after test set's validation, or the mobile reported good frames (CAT_TYPE 00) but actually validated by the test set as bad frame.
Frames Tested - the number of frames actually tested. If the Confidence Level is off, the Frames Tested equals the specified Max Frame Count when the FER measurement completes.
Max Frame Count - the maximum number of frames to be tested.
Eb/Nt - the ratio of energy per bit of the F-Traffic/F-FCH to noise power spectral density. It is displayed when AWGN power is on.
0 bps Frames Tested - the number of frames generated as 0 bps frames by the test set. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8) and Eighth Rate non- critical frames (see Rate 1/8 Non-Critical Frame Ratio ) are generated on the forward link.
0 bps Frame Errors - the number of frames not reported as Forward Erasures (having any CAT_TYPE other than `11') by the MS during the FER and the Eighth rate non-critical frame is generated on the forward link. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8).
Cur Ack Mask - the current ACK Mask for F-FCH.
FER (0 bps) - the ratio of the 0 bps Frame Errors Result to the 0 bps frames tested result when system type is IS-2000, and radio configuration is (Fwd11, Rvs8) and an Eighth rate non-critical frame is generated on the forward link.
PCG<n> Good Frames - the number of frames that correctly ACKed by the MS in each PCGs from 0 to 15. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8).
PCG<n> Bad Frames - the number of bad frames that ACKed by the MS in each PCGs from 0 to 15. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8).
FER (PCG 0-<FER First PCG Index>) - one minus that the ratio of the number of frames ACKed by the MS correctly between PCG 0 and PCG <FER First PCG Index> to the total number of frames tested. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8).
FER (PCG 0-<FER Second PCG Index>) - one minus that the ratio of the number of frames ACKed by the MS correctly between PCG 0 and PCG <FER Second PCG Index> to the total number of frames tested. It's only available when IS- 2000 radio configuration is (Fwd11, Rvs8).

A typical display is shown below:

FER Measurement Results Specific to Lab Application

In addition to the measurement results listed above, the FER measurements also provides the following results in the Lab Application :

Ext AWGN Power (dBm/1.23 MHz) - an external AWGN value is displayed whenever the test set is configured to generate faded forward channel signals with AWGN. This setup requires a rear panel digital bus connection to an external PC running Baseband Studio software. See Digital Bus Information for details.

When the test set is configured to work in FPC mode '000' (see FER Measurement Specific to the Lab Application ), the following additional results are available:

Eb/Nt Pass/Fail - indicates whether the measurement passed or failed the specified Eb/Nt Limit during the FER measurement. If the average Eb/Nt is no greater than the specified Eb/Nt Limit, the Eb/Nt Pass result is returned to indicate that the mobile station has passed the Eb/Nt limit for the FER test. Providing the Eb/Nt Pass/Fail result in addition to the FER confidence Pass/Fail result can help you identify which one leads to the final FER failure.
Average Eb/Nt - indicates the average Eb/Nt of the F-FCH during the FER measurement in FPC mode 000. This result is updated every 50 frames (~ 1 second) on the front panel display to indicate the most recent 50 frames average Eb/Nt result. When the measurement completes, the total average Eb/Nt during the FER measurement is returned.
Maximum Eb/Nt - indicates the maximum Eb/Nt of the F-FCH during the FER measurement in FPC mode 000.
Minimum Eb/Nt - indicates the minimum Eb/Nt of the F-FCH during the FER measurement in FPC mode 000.

A typical display for FER measurement in FPC mode 000 is shown below:

Input Signal Requirements

The FER measurement meets or exceeds specifications when the following requirements are met:

The frequency of the signal being measured must be in the range of 412 MHz to 483 MHz, 800 MHz to 960 MHz, or 1.7 GHz to 2.0 GHz, and within 100 kHz of the expected frequency.
The signal level into the test set's RF IN/OUT connector must be in the range of -25 dBm to +37 dBm, and within +/- 9 dB of the expected input power (see RFANalyzer:AUTO:POWer[:SELected]? to meet specifications.

Key TIA/EIA-98-E Tests Using FER

Testing 3.4.1 Demodulation of Forward Traffic Channel in AWGN
Testing 3.5.1 Receiver Sensitivity and Dynamic Range
The standard fading tests can be made when the test set is configured to work with the N5115A fading solution. For details, refer to:
E5515C Fading Solution Application Guide

It can be downloaded at http://cp.literature.keysight.com/litweb/pdf/1000-1894.pdf .