TD-LTE base station performance testing techniques based on HARQ

0 Introduction

LTE (Long Term Evolution) technology is the main direction of the evolution of the third generation of mobile communications. As an advanced technology, the LTE system has great advantages in improving peak data rate, cell edge rate, spectrum utilization, control plane and user plane latency, and reducing operation and network construction costs. At the same time, the LTE system can coexist with the existing system (2G/2.5G/3G) and achieve smooth evolution. The LTE

system is divided into frequency division duplex (FDD) and time division duplex (TDD) according to the duplex mode. Among them, the LTE-TDD standard has many advantages over the FDD standard, such as flexible spectrum utilization and support for asymmetric services. It is an international standard promoted by the Chinese communications industry.

System throughput is an important indicator to measure the comprehensive performance of TD-LTE base stations. The throughput test requires real-time feedback and dynamic adjustment between the base station (eNB) and the test instrument (simulated UE). The test instrument not only needs to be able to generate TD-LTE uplink signals that meet the 3GPP standard, but also needs to simulate the corresponding channel fading model and adjust the coding redundancy factor of the transmitted signal in real time according to the ACK/NACK instructions sent by the base station to simulate the real communication environment. This solution uses Agilent's baseband signal generator and channel emulator N5106A PXB or Agilent's latest signal generator N5182B/N5172B as the test platform. The TD-LTE test signal is generated by the Real-time version of Signal Studio N7625 for LTE TDD and sent to the base station for decoding after channel fading, so that the base station throughput can be counted.

1 HARQ Test Principle

1.1 Uplink HARQ Method
The LTE system will use synchronous non-adaptive HARQ technology in the uplink. Although asynchronous adaptive HARQ technology has higher scheduling flexibility than synchronous non-adaptive technology, the latter requires less signaling overhead. Due to the complexity of the uplink, the interference from users in other cells is uncertain, so the base station cannot accurately estimate the actual signal-to-interference ratio (SINR) value of each user. Due to the inaccuracy of SINR values, the uplink selection of modulation and coding mode (MCS) is not accurate enough, so the HARQ technology is more relied on to ensure the performance of the system. Therefore, the average transmission number of the uplink will be higher than that of the downlink. Therefore, considering the overhead of control signaling, synchronous non-adaptive HARQ technology is used in the uplink.

1.2 Uplink HARQ timing
The uplink and downlink signals of the LTE TDD standard are staggered in the time domain, so its HARQ timing mapping relationship is more complicated than that of FDD. According to 3GPP TS 36.213, under different UL/DL Configurations of the TDD standard, the downlink subframe only sends ACK/NACK instructions at the specified position, and the ACK/NACK instructions sent at each position correspond to a specific uplink subframe signal, as shown in Table 1.

Table 1 TD-LTE HARQ uplink and downlink subframe mapping relationship

Uploaded on 2013-8-19 11:12:44

Download attachment (33.26 KB)

 

3GPP TS 36.141 stipulates that performance testing only needs to be performed under configuration 1, so the timing diagram for configuration 1 can be obtained according to the description in Table 1:

Uploaded on 2013-8-19 11:12:42

Download attachment (58.43 KB)

 

Figure 1 HARQ timing diagram for configuration 1

In configuration 1, the uplink only sends uplink signals in subframes 2, 3, 7 and 8; the base station sends ACK/NACK instructions in subframes 1, 4, 6 and 9 for the downlink. The instruction object and retransmission position relationship are shown in Figure 1.

2 Test platform

2.1 Hardware platform
The purpose of performance testing is to simulate the system throughput in the actual environment, so the base station and the test instrument need to be jointly debugged. The hardware test platform includes: TD-LTE eNB base station supporting 2-antenna reception, Agilent baseband signal generator and channel emulator PXB, Agilent vector signal generator MXG (mainly used for up-conversion) and a four-channel oscilloscope (for system debugging). The test system structure is as follows:

Uploaded on 2013-8-19 11:12:42

Download attachment (79.2 KB)

 

Figure 2 N5106 PXB test system

PXB generates TD-LTE uplink signals in real time and after fading under a specific channel model, the output baseband I/Q signals are up-converted by MXG and sent to the two receiving antennas of the base station. The base station demodulates and decodes the received RF signals and transmits the feedback results (ACK/NACK instructions) back to PXB in RS232C serial communication mode. PXB adjusts the RV factor in real time according to the ACK/NACK instructions to resend the data packet or choose to abandon the current data packet (when the eNB sends an ACK signal or the maximum number of retransmissions has been reached). Finally, the base station obtains the system throughput by statistics.

If the test environment really wants to use an external channel emulator, you only need to use the Agilent N5182B/N5172B RF signal generator to complete the test of the above system. The typical measurement system is shown in Figure 3.

Uploaded on 2013-8-19 11:12:42

Download attachment (43.35 KB)

 

Figure 3 N5182B/72B MXG-B test system

2.2 Software platform
PXB or N5182B/72B generates specific reference test signals through Signal Studio N7625B-WFP for LTE TDD software and adjusts the coding redundancy factor in real time. [page]

Uploaded on 2013-8-19 11:12:43

Download attachment (50.06 KB)

 

Figure 4 Uplink signal configuration

Uploaded on 2013-8-19 11:12:43

Download attachment (5.84 KB)

 

Figure 5 HARQ setting

2.3 Feedback signal format
The ACK/NACK signal sent by the base station to PXB or N5182B/72B is encoded in the data format of RS232C serial communication. PXB or N5182B/72B decodes the ACK/NACK value according to the same coding rate and format. The feedback signal consists of 8 bits, 1 start bit, 1 stop bit, and no parity bit. The specific data format is shown in Table 2.

Table 2 ACK/NACK encoding format (LSB)

Uploaded on 2013-8-19 11:12:44

Download attachment (26.76 KB)

 

3 Test results

Build the test system and configure the software platform according to the test structure diagram, start the base station and PXB or N5182B/72B, and observe the test results on the oscilloscope and base station control terminal.

Uploaded on 2013-8-19 11:12:44

Download attachment (274.02 KB)

 

Figure 6 PXB or N5182B/72B real-time response

Figure 6 shows the real-time response of PXB or N5182B/72B observed on the oscilloscope. Channels 1, 2, 3 and 4 are the uplink signal frame header, ACK/NACK instruction sequence, uplink signal I/Q data and ACK/NACK response of PXB or N5182B/72B (high level is ACK, low level is NACK). As shown in the figure, the HARQ timing response is completely consistent with the standard protocol. The

base station performs diversity reception and demodulation on the uplink RF signal, and then judges the reception result through CRC check, and finally obtains the system throughput under a specific fading model.

According to 3GPP TS 36.141, PUSCH, 20MHz bandwidth signal is selected as the test case. Under 2 receiving antennas and Normal CP, the first 10 cases listed are tested in turn, and the results are as follows:

Table 3 PUSCH, PUSCH performance test results under 20MHz bandwidth

Uploaded on 2013-8-19 11:12:45

Download attachment (33.65 KB)

 

4 Conclusion

The results show that the system fully meets the standard test requirements. The test process is transparent and the results are intuitive and reliable. At the same time, the test system can achieve 1×2, 2×2, 2×4 and 4×2 MIMO configurations through software configuration without adding any hardware configuration, thereby realizing base station HARQ, Timing Adjustment and PUCCH performance testing, which is an ideal platform for TD-LTE base station performance testing.