802.11g EVM and BER-PER Performance

Introduction

WLAN_80211g_prj design examples are described in this chapter.

• WLAN_80211g_OFDM_TxEVM.dsn: measures error vector magnitude and relative constellation error and tests the transmit modulation accuracy for an OFDM signal.
• WLAN_80211g_CCK_TxEVM.dsn: measures error vector magnitude and relative constellation error and tests the transmit modulation accuracy for a CCK signal.
• WLAN_80211g_OFDM_36Mbps_Fading_System.dsn: BER and PER performance for 36 Mbps systems on a fading channel.
• WLAN_80211g_CCK_11Mbps_AWGN_System.dsn: BER and PER performance for 11 Mbps systems with CCK modulation on an AWGN channel.

Error Vector Magnitude and Relative Constellation Error Measurements

WLAN_80211g_OFDM_TxEVM.dsn

Features

• IEEE 802.11g configurable signal source, adjustable data rate
• Adjustable sample rate
• Constellation display
• Integrated RF section

Description

This design tests IEEE 802.11g transmit modulation accuracy and transmitter constellation error by measuring the EVM. The schematic for this design is shown in the following figure.

WLAN_80211g_OFDM_TxEVM.dsn Schematic

Measurements in this design are based on IEEE Standard 802.11a-1999 section 17.3.9.6. The transmit modulation accuracy test must be performed by instrumentation capable of converting the transmitted signal into a stream of complex samples at 20 Msamples per second or more, with sufficient accuracy in terms of I/Q arm amplitude and phase balance, dc offsets, phase noise, and so on. A possible embodiment of such a setup is converting the signal to a low IF frequency with a microwave synthesizer, sampling the signal with a digital oscilloscope and decomposing it digitally into quadrature components.

The sampled signal must be processed in a manner similar to an actual receiver, according to the following, or equivalent steps:

• Start of frame must be detected.
• Transition from short sequences to channel estimation sequences must be detected, and fine timing (with one sample resolution) must be established.
• Coarse and fine frequency offsets must be estimated.
• The packet must be de-rotated according to estimated frequency offset.
• The complex channel response coefficients must be estimated for each subcarrier.
• For each data OFDM symbol: transform the symbol into subcarrier received values, estimate the phase from the pilot subcarriers, de-rotate the subcarrier values according to estimated phase, and divide each subcarrier value with a complex estimated channel response coefficient.
• For each data-carrying subcarrier, find the closest constellation point and calculate the Euclidean distance from it.
• Calculate the RMS average of all errors in a packet:
  
  where
  L _P is the length of the packet
  N _f is the number of frames for the measurement
  ( I ₀ ( i, j, k ), Q ₀ ( i, j, k )) denotes the ideal symbol point of the i th frame, j th OFDM symbol of the frame, k th subcarrier of the OFDM symbol in the complex plane
  ( I ( i, j, k ), Q ( i, j, k )) denotes the observed point of the ith frame, jth OFDM symbol of the frame, k th subcarrier of the OFDM symbol in the complex plane (see the following figure).

Constellation Error

P ₀ is the average power of the constellation.
The vector error on a phase plane is shown in the following figure.

EVM and Relative Constellation Error of 54 Mbps

The test must be performed over at least 20 frames ( N _f ) and the RMS average must be taken. The packets under test must be at least 16 OFDM symbols long. Random data must be used for the symbols.

The EVM and relative constellation RMS error, averaged over subcarriers, OFDM frames, and packets, cannot exceed a data-rate dependent value according to the following table.

Allowed EVM and Relative Constellation Error

Data Rate (Mbps)	Relative Constellation Error (dB)	EVM (% RMS)
6	-5	56.2
9	-8	39.8
12	-10	31.6
18	-13	22.3
24	-16	15.8
36	-19	11.2
48	-22	7.9
54	-25	5.6

Simulation Results

Simulation results displayed in WLAN_80211g_TxEVM.dds are shown in the preceding figure for EVM and relative constellation error of 54 Mbps. The EVM is less than 0.6%; the constellation error is approximately -45dB which is much smaller than the specification requirements given in the preceding table.

Benchmark

• Hardware platform: Pentium III 450 MHz, 512 MB memory
• Software platform: Windows NT 4.0 Workstation, ADS 2001
• Simulation time: approximately 30 minutes

References

1. IEEE Standard 802.11a-1999, "Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: High-speed Physical Layer in the 5 GHz Band," 1999.

Error Vector Measurement for a CCK Signal

WLAN_80211g_CCK_TxEVM.dsn

Features

• IEEE 802.11g configurable signal source, adjustable data rate
• Adjustable sample rate
• CCK modulation

Description

This design tests IEEE 802.11g transmit modulation accuracy by measuring the EVM. The schematic for this design is shown in the following figure.

WLAN_80211g_CCK_TxEVM.dsn Schematic

Measurements in this design are based on IEEE Standard 802.11b-1999 section 18.4.7.8. The transmit modulation accuracy requirement for the high rate PHY is based on the difference between the actual transmitted waveform and the ideal signal waveform. Modulation accuracy is determined by measuring the peak vector error magnitude during each chip period. Worst-case vector error magnitude cannot exceed 0.35 for the normalized sampled chip data. The ideal complex I and Q constellation points associated with DQPSK modulation, (0.707,0.707), (0.707, -0.707), (-0.707, 0.707), (-0.707, -0.707), will be used as the reference. These measurements are from baseband I and Q sampled data after recovery through a reference receiver system. The measurement example is shown in the following figure.

Modulation Accuracy Measurement Example

Simulation Results

Simulation results displayed in WLAN_80211g_TxEVM.dds are shown in the following figure. The EVM results smaller than the specification requirements.

EVM Results

Benchmark

• Hardware platform: Pentium III 450 MHz, 512 MB memory
• Software platform: Windows NT 4.0 Workstation, ADS 2002c
• Simulation time: approximately 4 minutes

References

1. IEEE Standard 802.11b-1999, "Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: High-speed Physical Layer Extension in the 2.4 GHz Band," 1999.

BER and PER Performance, Fading Channel 36 Mbps

WLAN_80211g_OFDM_36Mbps_Fading_System.dsn

Features

• Data rate=36Mbps, coding rate=3/4, modulation=16-QAM, velocity=0 km/hr
• Length and Order parameter default settings = 512 and 7, respectively
• BER and PER vs. Eb/N0 on fading channel curves displayed

Description

This design shows system performance with 36 Mbps data rate and channel coding on fading channel. A burst length of 512 bytes is simulated.

The top-level schematic for this design is shown in the following figure. SignalSource parameters are contained in Signal_Generation_VARs; Noise, Receiver, and BERPER parameters are contained in RF_Channel_Measurement_VARs.

WLAN_80211g_OFDM_36Mbps_Fading_System.dsn Schematic

According to reference [2], five model types have been designed. Model A, an 18-tap fading channel corresponding to a typical office environment for NLOS conditions and a 50ns average rms delay spread, is used in this example. In order to reduce the number of taps needed, the time spacing is non-uniform; for shorter delays, a more dense spacing is used. The average power declines exponentially with time. For Model A, all taps have Rayleigh statistics. The characteristics of this model are listed in the following table.

Model A Characteristics

Tap Number	Delay (ns)	Average Relative Power (dB)	Ricean K	Doppler Spectrum
1	0	0.0	0	Class
2	10	-0.9	0	Class
3	20	-1.7	0	Class
4	30	-2.6	0	Class
5	40	-3.5	0	Class
6	50	-4.3	0	Class
7	60	-5.2	0	Class
8	70	-6.1	0	Class
9	80	-6.9	0	Class
10	90	-7.8	0	Class
11	110	-4.7	0	Class
12	140	-7.3	0	Class
13	170	-9.9	0	Class
14	200	-12.5	0	Class
15	240	-13.7	0	Class
16	290	-18.0	0	Class
17	340	-22.4	0	Class
18	390	-26.7	0	Class

Simulation Results

Simulation results are shown in the following two figures.

For PER performance, the WLAN_OFDM_80211g_36Mbps_Fading_System.dsn is approximately 2 dB worse than that of WLAN_80211g_24Mbps_Fading_System.dsn.

Fading Channel BER Performance

Fading Channel PER Performance

Benchmark

• Hardware platform: Pentium III, 500 MHz, 512 MB memory
• Software platform: Windows NT 4.0, ADS 2002
• Data points: Eb/N0 values is set from 10 to 15 dB
• Simulation time: 50 hours

References

1. IEEE Standard 802.11a-1999, "Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: High-speed Physical Layer in the 5 GHz Band," 1999.
2. Channel Models for HIPERLAN/2 in Different Indoor Scenarios, ETSI EP BRAN 3ER1085B 30 March 1998.

BER and PER Performance, AWGN Channel 11 Mbps for CCK Signal

WLAN_80211g_CCK_11Mbps_AWGN_System.dsn

Features

• Data rate = 11Mbps
• Modulation = CCK
• Carrier frequency offset between transmitter and receiver = 50 kHz
• BER and PER vs. Eb/N0 on AWGN channel curves are displayed

Description

This design shows system performance of 802.11g with 11Mbps data rate and CCK modulation on an AWGN channel. A burst length of 500 bytes is simulated.

The top-level schematic is shown in the following figure. This design contains SignalSource, Noise, Receiver, and BERPER subnetworks. SignalSource parameters are contained in the Signal_Generation_VARs; Noise and BERPER parameters are contained in the RF_Channel_Measurement_VARs; and, Receiver parameters are contained in the Receiver_VARs.

WLAN_80211b_11Mbps_AWGN_System.dsn Schematic

The SignalSource subnetwork (see the following figure), generates a signal source based on user settings.

SignalSource Subnetwork Schematic

The Receiver subnetwork (see the following figure) receives an RF signal and demodulates the signal as bit streams; it also detects the start of frame and completes the transition from received sequences to frequency offset estimation sequences, estimates the frequency offset caused by carrier differences between transmitter and receiver. A decision feedback equalizer is implemented to equalize the received signal and remove the fixed rotation caused by frequency offset. The equalized signal is then fed into the CCK demodulator and demodulated into bit streams.

Receiver Subnetwork Schematic

The BERPER subnetwork (see the following figure) measures system BER and PER.

BERPER Subnetwork Schematic

Simulation Results

Simulation results are shown in the following figure.

Reference data points are shown in page Equations.

Simulation Results

Benchmark

• Hardware platform: Pentium IV, 1.8 GHz, 512 MB memory
• Software platform: Windows XP, ADS 2003A
• Data points: Eb/N0 values is set from 7 to 10 dB
• Simulation time: 1.5 hours

References

1. IEEE Standard 802.11a-1999, "Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: High-speed Physical Layer in the 5 GHz Band," 1999.
2. Intersil, "Direct Sequence Spread Spectrum Baseband Processor with Rake Receiver and Equalizer," Data Sheet, FN4856.2, December 2001.