2005 EMPLOYEE S REPORT ON ACTIVITIES AND ACCOMPLISHMENTS
Document Sample


2005 EMPLOYEE’S REPORT ON ACTIVITIES AND ACCOMPLISHMENTS
Name: Dragan Samardzija Cost Center: BL10009650 SRD: 6/12/00 Date: September 2005
I have been working on the following projects:
1. MIMO HSDPA demonstration and outdoor measurements for TNO
It was externally funded by TNO. This project consisted of the following phases.
1. 4x4 MIMO indoor demonstration - The first trip was in November 2004 to TNO, Delft, the Netherlands. 4x4
MIMO HSDPA system was presented at the TNO-organized event for their largest customer, KPN. The
demonstration attracted considerable interest from the KPN visitor. Enabling the 4x4 MIMO equalizer at the
terminal side and the OneBTS MIMO transmitter was the major development work prior to the demonstration.
Furthermore, significant effort was put into organizing the shipment, packaging and installation of the equipment.
Peter Bosch helped me during this first trip and demonstration.
2. 4x4 MIMO outdoor measurements - The second trip was in March 2005, and consisted of two weeks of outdoor
measurements. The measurement setup consisted of (i) OneBTS as the MIMO transmitter, (ii) MIMO terminal
with the MIMO equalizer ASIC and turbo decoder ASIC, and (iii) the MIMO sounder that was recording received
data for post processing. During the months of December ‟04, January and February ‟05 I have invested
considerable time (i) to integrate wireless synchronization, (ii) to integrate AGC, (iii) to improve the terminal
sensitivity (resulting in 9.6 Mbps rates at –95 dBm, in static conditions) by integrating different LNAs and
improving both the RF front-end and the digital baseband (i.e., ADs and FPGA-based processing), (iv) to enable 4
10 W PAs, (v) to calibrate the MIMO sounder and recording system. During the measurements we encountered a
number of problems Firstly, there was an adjacent commercially active UMTS carrier that was greatly exceeding
our signal strength (up to 20 dB) when distances from our transmitter exceeded couple of hundred of meters. This
resulted in our receiver AGC detecting adjacent signal, and attenuating the desired signal, consequently
diminishing the performance. Secondly, we concluded that the MIMO equalizer failed to track the channel
fluctuations in presence of low SNR (under 3 dB) and significant delay spread (when exceeding 1 usec, rms). The
sounder was performing without any problem, collecting data at different locations in Delft and Eindhoven. Sape
Mullender helped me during the measurements and data collection.
3. Postprocessing of the sounder data - Using the collected sounder measurements, during March and April ‟05 I
extracted the following quantities: (i) wideband MIMO channel impulse response, (ii) SNR, (iii) estimation SNR,
(iv) rms of the delay spread, (v) optimal estimation period, (vi) MIMO ASIC throughput and (vii) wideband
MIMO channel capacity. There were more that 700 measurement locations. The results were delivered to TNO,
and presented by TNO in the Netherlands, in May ‟05 [1]. Sape Mullander helped me with the data processing.
2. Dynamic spectrum management
Most of the work is done with our summer intern Omer Ileri, Tod Sizer and Professor Narayan Mandayam. During
the months of June through September ‟05 I was supervising and advising Omer Ileri, on a daily basis.
We have come up with a concise mathematical framework that quantifies user service satisfaction for the given price
and offered rate. Based on that we formulated a scenario where multiple service providers are competing for
customers and available spectrum resources.
Furthermore, the proposed competition and spectrum allocation is to be supervised by a central authority (namely
Spectrum Policy Server (SPS)). The results are published as a TM and it is accepted for IEEE Symposium on New
Frontiers in Dynamic Spectrum Access, Baltimore 2005 [2].
The results clearly point to (i) great benefits of competition between the operators (increasing users‟ acceptance of
the offered services), (ii) great improvements in spectrum utilization (i.e., users are enticed to use more spectrum),
(iii) profits of the individual operators are lowered (versus the case without competition), and (iv) ability to manage
the spectrum as a profitable entity while increasing spectrum utilization (i.e., SPS can be profitable).
Currently I am working with Omer Ileri, now at WINLAB, Rutgers University, and Professor Mandayam on
practical solutions that would enable this scheme
Furthermore, I have outlined “wireless franchise” that is based on the above ideas and I have hopes that it will result
in a new scheme how spectrum and wireless access can be managed in the future. I would like to develop this idea
further during the next year.
3. Pilot recovery and generation for RadioStar
This work was done June through September ‟05. The idea was to synchronize to one carrier coming out of a BTS,
and then at the remote side of the RadioStar system, to generate pilots on other, non-existing carriers in order to
support inter-carrier handoff. It consisted of the following phases.
DOCUMENTATION – SEE REVERSE
Dragan Samardzija – 2
Build a synchronization scheme based on pilot recovery – It was implemented on FPGA in the RadioStar base
1.
station interface (BSI). Major challenge was to fit the design into the space-limited FPGA with tight timing
constraints.
2. Build a pilot generation – It was implemented of FPGA in the RadioStar remote radio head (RRH). I had to build
(i) pilot generator, (ii) interpolation filter based on the IS-95 specs, (iii) digital modulator that places the pilot on
multiple different IS-95 carriers.
3. Enable synchronization of the RRH pilot generation with the BSI pilot synchronization via the Gigabit
Ethernet – I enabled pilot recovery to send synchronization information at the BSI side and use it at the RRH
side. Xiang Ma developed the microprocessor and Ethernet part of the solution.
Recently, the solution was fully integrated with the RadioStar system, and has passed a number of tests.
4. Network MIMO
The intent is to demonstrate 4 access points (i.e., BTSs) performing generalized beamforming that zeros interference
between signals sent to two independent terminals.
I am guiding work that is done by Clark Woodworth, Sue Walker and Howard Huang and myself.
We have completed the following blocks: (i) transmitter signal generation (antenna pilot, dedicated pilot and data),
(ii) generalized beamformer, (iii) beamformer coefficient calculation, (iv) channel estimator, and (v) CSI feedback.
Currently I am integrating the individual blocks. I expect the whole system to be integrated sometimes in November
„05. Thus, the first tests will be conducted in December „05 with demonstrations in early ‟06.
If this stage is successful the following steps may be: (i) wireless CSI feedback instead of the existing wired one, and
(ii) using RadioStar to distribute signals from the central processor via the Ethernet instead of the current coaxial
cable distribution.
5. Application of MIMO techniques to sensing of cardiopulmonary activity
This is joint project with the University of Hawaii (Olga and Victor Lubecke and Anders Host Madsen) and Bell
Labs.
In November ‟04 we were informed that this project was to be funded by NSF. I was one of the authors of the
proposal and Co-PI.
I came up with a number of possible schemes how to use multiple antennas to enhance sensing of cardiopulmonary
activity. The schemes and improvements are presented in [3] and [4].
Using the project fund I have visited university of Hawaii and met with a number professors and students. One of the
students spent this summer working on the RadioStar project, while being funded by the University of Hawaii.
We expect to get more students in upcoming months and certain funds for the equipment (to be paid by the NSF
grant).
Publications
Presentations:
[1] “Measuring and Predicting UMTS MIMO Performance in Outdoor Environments”, presented by TNO, the Netherlands,
May 2005.
Conference publications:
[2] O. Ileri, D. Samardzija, T. Sizer, N. Mandayam, “Demand Responsive Pricing and Competitive Spectrum Allocation via a
Spectrum Server”, IEEE Symposium on New Frontiers in Dynamic Spectrum Access, Baltimore 2005.
[3] D. Samardzija, O. Boric-Lubecke, A. Host-Madsen, V. M. Lubecke, T. Sizer, A. D. Droitcour, G. T. A. Kovacs, “Applications of
MIMO Techniques to Sensing of Cardiopulmonary Activity”, IEEE/ACES International Conference on Wireless
Communications and Applied Computational Electromagnetics, Honolulu, 2005.
[4] O. Boric-Lubecke, V. M. Lubecke, A. Host-Madsen, D. Samardzija, K. Cheung , “Doppler Radar Sensing of Multiple Subjects in
Single and Multiple Antenna Systems”, 7th International Conference on Telecommunications in Modern Satellite, Cable and
Broadcasting Services (TELSIKS), Nis, 2005.
[5] D. Chizhik, J. Ling, D. Samardzija, R. A. Valenzuela, “Spatial and Polarization Characterization of MIMO Channels in Rural
Environment”, the IEEE Vehicular Technology Conference (VTC), Stockholm, 2005.
[6] J. Ling, D. Chizhik, D. Samardzija, R. A. Valenzuela, "Wideband and MIMO Measurements in Wooded and Open Areas", IEEE
Conference on Antennas and Propagation, Washington, 2005.
[7] D. Samardzija, N. Mandayam, “Unquantized and Uncoded Channel State Information Feedback on Wireless Channels”, IEEE
WCNC, New Orleans 2005.
Journal publications:
[8] D. Samardzija, A. Lozano, C. Papadias, "Design and Experimental Validation of MIMO Multiuser Detection for Downlink
Packet Data", EURASIP Journal on Applied Signal Processing, no. 11, pp. 1769-1778, July 2005.
[9] D. Samardzija, N. Mandayam, D. Chizhik, "Adaptive Transmitter Optimization in Multiuser Multiantenna Systems: Theoretical
Limits, Effect of Delays and Performance Enhancements", EURASIP Journal on Wireless Communications and Networking,
Issue 3, August 2005.
Signature: Dragan Samardzija
Date: 10/05/2005
Related docs
Get documents about "