Experiments with an 802.11n Radio Testbed July 2005 doc.: IEEE 802.11-05/11-05-0701-02-000n July 2005 Experiments with an 802.11n Radio Testbed Date: 2005-07-19 Authors: Notice: This document has been prepared to assist IEEE 802.11. It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE 802.11. Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures <http:// ieee802.org/guides/bylaws/sb-bylaws.pdf>, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair <stuart.kerry@philips.com> as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE 802.11 Working Group. If you have questions, contact the IEEE Patent Committee Administrator at <patcom@ieee.org>. David Browne, UCLA UnWiReD Lab David Browne, UCLA UnWiReD Lab

July 2005 doc.: IEEE 802.11-05/11-05-0701-02-000n July 2005 Abstract Present a software defined MIMO-OFDM testbed with 802.11n functionality. Present a channel sounding scheme that can be integrated into 802.11n testbeds. Demo simultaneous channel sounding and 802.11n measurements on the testbed via internet. Present results from a measurement campaign in which a comparison was made between proposed 802.11n modem/codec schemes. David Browne, UCLA UnWiReD Lab David Browne, UCLA UnWiReD Lab

Experiments with an 802.11n Radio Testbed July 2005 Experiments with an 802.11n Radio Testbed U n W i R e D L a b UCLA Wireless Research and Development (in collaboration with STMicroelectronics) Michael Fitz David Browne, Weijun Zhu David Browne, UCLA UnWiReD Lab

July 2005 Outline Present a software defined MIMO-OFDM testbed with 802.11n functionality. Present a channel sounding scheme that can be integrated into 802.11n testbeds. Demo simultaneous channel sounding and 802.11n measurements on the testbed via internet. Present results from a measurement campaign. David Browne, UCLA UnWiReD Lab

MIMO-OFDM Radio Testbed July 2005 MIMO-OFDM Radio Testbed David Browne, UCLA UnWiReD Lab

MIMO-OFDM Radio Testbed July 2005 MIMO-OFDM Radio Testbed David Browne, UCLA UnWiReD Lab

MIMO-OFDM Radio Testbed July 2005 MIMO-OFDM Radio Testbed Radios digital baseband 4 x 4 MIMO fc = 2.4GHz 20MHz bandwidth 120Mbps throughput 3% EVM internet control David Browne, UCLA UnWiReD Lab

MIMO-OFDM Radio Testbed July 2005 MIMO-OFDM Radio Testbed Robots spatial positioning 70cm x 70cm grid 1.59mm precision LEGO internet control Antenna Arrays variable geometry dipoles, patches compact MIMO arrays David Browne, UCLA UnWiReD Lab

MIMO-OFDM Radio Testbed July 2005 MIMO-OFDM Radio Testbed Fair comparison of MIMO-OFDM signaling schemes Multipath distortion Phase noise Frequency offset Non-linear amplifiers Antenna mutual coupling ... Method: simultaneous channel sounding and data transmission David Browne, UCLA UnWiReD Lab

Field Test - Super Frame July 2005 Field Test - Super Frame Superframe Structure: Up to 800us long. One channel sounding packet. Data packets with possibly different coding and modulation schemes. Silence period between packets. David Browne, UCLA UnWiReD Lab

Field Test - Super Frame July 2005 Field Test - Super Frame Assumption: Channel characteristics does not change significantly over the super frame. Advantage: Possible to characterize data packet’s performance with the knowledge of channel environment. Each scheme’s frame is self contained: synchronization fields (similar to 802.11a) preamble payload information packet David Browne, UCLA UnWiReD Lab

Field Test - General Frame Structure July 2005 Field Test - General Frame Structure 802.11a Preamble for timing and frequency estimation. Flexible structure for testing different MIMO-OFDM schemes simultaneously. Implemented algorithms include Spatial multiplexing and space-time blocking coding, LDPC/BCC/Space-time trellis coding, etc. Many key algorithms related to the TGn proposals have been implemented and tested. David Browne, UCLA UnWiReD Lab

Channel Sounding Packets July 2005 Channel Sounding Packets David Browne, UCLA UnWiReD Lab

July 2005 Outline Present a software defined MIMO-OFDM testbed with 802.11n functionality. Demo simultaneous channel sounding and 802.11n measurements on the testbed via internet. Present a channel sounding scheme that can be integrated into 802.11n testbeds. Present results from a measurement campaign. David Browne, UCLA UnWiReD Lab

July 2005 David Browne, UCLA UnWiReD Lab

July 2005 Live Demo David Browne, UCLA UnWiReD Lab

July 2005 Outline Present a software defined MIMO-OFDM testbed with 802.11n functionality. Demo simultaneous channel sounding and 802.11n measurements on the testbed via internet. Present a channel sounding scheme that can be integrated into 802.11n testbeds. Present results from a measurement campaign. David Browne, UCLA UnWiReD Lab

Channel Sounding Popular Methods: Delay domain channel sounders July 2005 Channel Sounding Popular Methods: Delay domain channel sounders Correlation methods that estimate channel impulse response. Swept frequency channel sounders SISO hardware adapted to MIMO (virtual & switched arrays). Require time synchronization. Difficult to detect co-channel interference. Do not utilize .11n radio architecture. David Browne, UCLA UnWiReD Lab

Channel Sounding Proposed Scheme: July 2005 Channel Sounding Proposed Scheme: Space-Frequency method suited to MIMO-OFDM. Does not require time synchronization of TX and RX. Integrated into testbed packet scheme. Instantaneous channel capacity results. reflects hardware and channel distortion experienced by packets. Detects co-channel interference during a measurement. David Browne, UCLA UnWiReD Lab

Channel Sounding The space-frequency signaling scheme … July 2005 Channel Sounding The space-frequency signaling scheme … … and associated channel estimation algorithm. David Browne, UCLA UnWiReD Lab

Channel Sounding Transmitter (2 x 2 Example) July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Excitation Signal Synthesis… July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Excitation Signal Spectra… July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Space-Frequency Orthogonality … July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Excitation Signal Waveform Poor PAPR … July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Excitation Signal Waveform (good PAPR) … July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Receiver (2 x 2 Example) July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Orthogonal transformation by FFT… July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Noise detection … July 2005 David Browne, UCLA UnWiReD Lab

July 2005 Channel Sounding David Browne, UCLA UnWiReD Lab

Channel Sounding Signal detection … July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Channel estimation … July 2005 David Browne, UCLA UnWiReD Lab

Channel Sounding Interference detection … Nominal Signal July 2005 Channel Sounding Interference detection … Nominal Signal Nominal Noise Floor 802.11b Interference Bluetooth Inteference David Browne, UCLA UnWiReD Lab

Channel Sounding Advantages: July 2005 Channel Sounding Advantages: Space-Frequency orthogonallity suited to MIMO-OFDM. Signal periodicity → no time synchronization. Low PAPR → max range and linear RF operation. Utilizes 802.11n radio architecture. Detection of co-channel interference David Browne, UCLA UnWiReD Lab

System Benchtest on a 2x2 MIMO Channel Emulator… July 2005 System Benchtest on a 2x2 MIMO Channel Emulator… David Browne, UCLA UnWiReD Lab

Ideal Gaussian 2x2 MIMO Channel July 2005 Ideal Gaussian 2x2 MIMO Channel David Browne, UCLA UnWiReD Lab

Indoor Correlated Spatial Channel July 2005 Indoor Correlated Spatial Channel David Browne, UCLA UnWiReD Lab

Single Path Keyhole Channel July 2005 Single Path Keyhole Channel David Browne, UCLA UnWiReD Lab

July 2005 Outline Present a software defined MIMO-OFDM testbed with 802.11n functionality. Demo simultaneous channel sounding and 802.11n measurements on the testbed via internet. Present a channel sounding scheme that can be integrated into 802.11n testbeds. Present results from a measurement campaign. David Browne, UCLA UnWiReD Lab

Field Tests Scenario: Indoor Office Environment July 2005 Field Tests Scenario: Indoor Office Environment Line-of-Sight & Non-Line-of-Sight Variable TX power (-6dBm ~ 15dBm) Linear array of dipoles 0.5 λ antenna spacing 11 x 9 positioning grid with λ/2 grid spacing 10 time separated measurements per grid point David Browne, UCLA UnWiReD Lab

July 2005 Field Tests David Browne, UCLA UnWiReD Lab

July 2005 Capacity vs. SNR David Browne, UCLA UnWiReD Lab

July 2005 Capacity vs. SNR David Browne, UCLA UnWiReD Lab

Capacity vs. SNR >10 000 data measurement points. July 2005 Capacity vs. SNR >10 000 data measurement points. Error statistics binned by SNR (left) and Capacity (right). SNR does not expose trend we expect (too few points). Capacity plot better at exposing the error rate profile. David Browne, UCLA UnWiReD Lab

Capacity vs. SNR Capacity is better suited to characterizing MIMO July 2005 Capacity vs. SNR Capacity is better suited to characterizing MIMO performance in field measurements Accounts for both SNR and channel structure. Fewer degrees of freedom → fewer data points needed. David Browne, UCLA UnWiReD Lab

Field Test - Compared Schemes July 2005 Field Test - Compared Schemes Experimental Modem/Codec Parameters: 2 x 2 and 2 x 3 data signaling 4 x 4 channel sounding. WWise LDPC and BCC, Rate 1/2, 2/3, 5/6. 16QAM and 64QAM. MIMO Preamble based on Hadamard Sequence. Pilot Tones (Orthogonal w.r.t different TX channels). 1000 information bytes per data packet. David Browne, UCLA UnWiReD Lab

Field Test - Observation Perspectives July 2005 Field Test - Observation Perspectives LDPC vs. BCC LOS vs. NLOS 2RX vs. 3RX Plotting against capacity vs. SNR Hardware Limitation David Browne, UCLA UnWiReD Lab

Field Test Results - LDPC vs. BCC (1) July 2005 Field Test Results - LDPC vs. BCC (1) LDPC significantly outperforms BCC, especially at high rate. David Browne, UCLA UnWiReD Lab

Field Test Results - LDPC vs. BCC (2) July 2005 Field Test Results - LDPC vs. BCC (2) BCC shows a much higher error floor than LDPC. David Browne, UCLA UnWiReD Lab

Field Test Results - LOS vs. NLOS (1) July 2005 Field Test Results - LOS vs. NLOS (1) Comparison not obvious when plotting PER against capacity. David Browne, UCLA UnWiReD Lab

Field Test Results - LOS vs. NLOS (2) July 2005 Field Test Results - LOS vs. NLOS (2) LOS suffers performance degradation at higher SNR compared with NLOS except with the highest rate mode. David Browne, UCLA UnWiReD Lab

Field Test Results - 2 RX vs. 3 RX (1) July 2005 Field Test Results - 2 RX vs. 3 RX (1) Performance difference is significant when plotting against SNR. David Browne, UCLA UnWiReD Lab

Field Test Results - 2 RX vs. 3 RX (2) July 2005 Field Test Results - 2 RX vs. 3 RX (2) Performance is very close for most of test modes when plotting against the capacity. Indicating that capacity is a better choice of measure of channel. David Browne, UCLA UnWiReD Lab

Field Test Results - Hardware Limitation July 2005 Field Test Results - Hardware Limitation Maximum SNR achievable is about 32 ~ 34 dB on the testbed. David Browne, UCLA UnWiReD Lab

Field Test Conclusions July 2005 Field Test Conclusions Capacity better suited to characterizing MIMO performance in measurement campaigns than SNR . LDPC outperforms BCC. LOS performance degradation observed. Direct comparison of how close 802.11n performs compared to the achievable channel capacity. David Browne, UCLA UnWiReD Lab

July 2005 Conclusion Presented a software defined MIMO-OFDM testbed with 802.11n functionality. Presented a channel sounding scheme that can be integrated into 802.11n testbeds Demoed simultaneous channel sounding and 802.11n measurements on the testbed via internet. Presented results from a measurement campaign. David Browne, UCLA UnWiReD Lab

July 2005 References W. Zhu, D.W. Browne, M.P. Fitz, "An Open Access Wideband Multi-Antenna Wireless Testbed with Remote Control Capability", IEEE TridentCom, February 2005. D. W. Browne, W. Zhu, and M.P. Fitz, "A Signaling Scheme and Estimation Algorithm for Characterizing Frequency Selective MIMO Channels," IEEE Vehicular Technology Conference, May 2005. D. W. Browne, M. Manteghi, M. P. Fitz, Y. Rahmat-Samii, "Antenna Topology Impacts on Measured MIMO Capacity," IEEE Vehicular Technology Conference, September 2005. David Browne, UCLA UnWiReD Lab