Conformance Testing of LTE User Equipment

By: Sandy Fraser, Agilent Technologies
( 1 Nov 2009 )

The purpose of conformance testing is to ensure a minimum level of performance, and, as with all previous mobile radio systems, LTE (Long Term Evolution) will require the performance of an extensive list of tests. This list becomes longer as the number of variables to be tested increases.

The successful deployment of any mobile communications system depends on the compatibility and effective inter-working of the system's different elements. These comprise the UE (user equipment) and the network elements with which the UE has to interact – principally the basestation. Standards define the operation and minimum acceptable performance for UEs and basestations. Conformance testing ensures that products meet these standards. Doubling the number of possible modulation depths, for example, doubles the possible number of tests. Considering the possible combinations of resource allocations, modulation formats, frequency bands and mobility options – including CDMA2000, 2G and 3G handovers – LTE would seem to require a very long list of tests. However, the test regime has been designed to minimize test time, and the design of the LTE system with less UE states could reduce test time.

LTE TEST STRUCTURE.
The LTE conformance test structure generally follows closely that for UMTS: each test describes the purpose of the test; the test applicability, e.g. LTE TDD or FDD; minimum requirements, which equal the lowest acceptable performance level); the conditions under which the test should be performed; elements such as message contents, UE state and transmission properties; the procedures to execute the test; and the test pass/fail conditions.

In UMTS, each conformance test must be started exactly as specified by the conformance requirements. For LTE, the UE and the test equipment are only required to be set up in the correct starting condition for the tests. It may be possible to run several tests without tearing down radio bearers between tests, saving considerable time compared with UMTS tests. This is only possible if, in this example, the radio bearer is common – or can be modified – between consecutive conformance tests, but there should be considerable scope for time reduction.

The LTE conformance-test standards are split into two major documents: the three-part 36.521 deals with all the transmitter and receiver tests and RMM (radio resource management), while 36.523 deals with the signaling (protocol) tests.

USER-EQUIPMENT CONFORMANCE TESTS
The list of transmitter test cases includes all of the expected transmit measurements, which ensure a minimum level of interference generation that could affect other UEs or the network performance. These measurements include EVM (error-vector magnitude), transmit power, adjacent-channel-leakage ratio, spurious emissions, and intermodulation. Most of these tests can be performed prior to conformance with analysis equipment such as the Agilent MXA range of signal analyzers.

Receiver conformance tests include reference sensitivity, the minimum level at which the UE can receive and decode a signal, plus selectivity and blocking tests to ensure that the UE can detect signals intended for that UE in the presence of other LTE signals and noise. The receiver test cases are generally conducted in the presence of AWGN (additive white Gaussian noise), which is used to simulate a noisy environment similar to that of a real LTE system. Unlike 2G/3G, which includes tests for BER (bit-error rate), all LTE receiver tests are measured in terms of BLER (block-error rate).

Performance tests establish the capability of the UE under conditions more like the situations under which it will be used. For example, as well as the presence of AWGN as an interferer, the performance tests require the use of fading. Performance tests with a variety of channel conditions such as diversity and MIMO are defined in 36.521-1. Performance testing often requires the use of a closed-loop system where the UE’s responses to the conditions during the tests are part of the test set-up. For instance, the CQI (channel-quality indicator) is reported by the UE to the eNB and indicates the channel conditions. This CQI value can then be used by the eNB to control the amount of data sent to the UE in order to continually vary and maximize throughput according to the channel conditions.

CHANNEL-QUALITY-INDICATOR TESTS
The CQI tests are conducted in two parts. The first part records the CQI reported by the UE, which is then used to measure throughput. The test passes if the data-throughput rates match those for the given CQI value. This test requires an instrument with real-time protocol control, such as the E6620A wireless communications test set, which runs the Agilent LTE mobile test software (Figure 1). For LTE, the current conformance standards cover the minimum requirements and basic test needs. FRCs do not test the UE under real-world operating conditions. The RRM core specifications (36.133) are partially completed, with tests defined for mobility measurements – reference signal receive power and quality – and cell re-selection between LTE cells. The second phase will add tests for cell re-selection between LTE and 2G/3G plus RRC connection control.

For the transmitter, receiver and performance conformance tests to be conducted in a repeatable and comparable manner, 3GPP has defined a number of FRCs (fixed reference channels). There are many FRCs required for transmitter and receiver tests to simulate different situations, e.g. transmit diversity or MIMO, QPSK or 64QAM transmission with differing data rates. However FRCs cannot test a device under variable, real-world conditions.

Signaling conformance testing ensures that the UE correctly handles link-control information and data-flow control. The current standards include tests for handover procedures, paging and neighbor-cell reports, ciphering procedures, and correct operation of data transfer. Conformance testing will provide the minimum acceptable test level for LTE, to be augmented by additional characterization and performance testing.

Click here for the illustrations:

Figure 1