90 likes | 250 Vues
08/07/2010. Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [ Potential Channel Tracking Limitations in OFDM Mode] Date Submitted: [ July 2010 ] Source: [Charles Razzell] Company: [Maxim Integrated Products] E-Mail: [ crazzell@ieee.org ]
E N D
08/07/2010 Project: IEEE P802.15 Working Group for Wireless Personal Area Networks (WPANs) Submission Title: [ Potential Channel Tracking Limitations in OFDM Mode] Date Submitted: [ July 2010 ] Source:[Charles Razzell] Company: [Maxim Integrated Products] E-Mail: [ crazzell@ieee.org] Re: [ TG4g comments on LB51 (e.g., #1422)] Abstract: Purpose: Provide data to assess potential limitationsof the proposed OFDM pilot scheme Notice: This document has been prepared to assist the IEEE P802.15. 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 acknowledges and accepts that this contribution becomes the property of IEEE and may be made publicly available by P802.15. Slide 1
08/07/2010 Background • A specific OFDM pilot scheme has been proposed for 802.15.4g • This document seeks to estimate the limits of the channel disturbances that can be successfully tracked using the proposed pilot scheme • Decision directed tracking is not considered at this time. • Knowing these limitations should inform any discussion about the merits of the existing scheme. Slide 2
08/07/2010 Pilot Scheme for Option 2 Slide 3
08/07/2010 Pilot Scheme Overlaid with 8us 2-tap Channel 8us * C = 2.4km flight time difference. Slide 4
08/07/2010 Pilot Scheme Overlaid with 8us 2-tap Channel 8us * C = 2.4km flight time difference. Under-sampled Slide 5
08/07/2010 Pilot Scheme Overlaid with 8us 2-tap Channel 8us * C = 2.4km flight time difference. Slide 6
08/07/2010 Rate of change of channel • If the channel is quasi-stationary over 9 OFDM symbols, the additional frequency domain samples can be “safely” used without additional processing. • Assume we will allow 6 of rotation over 9 x 120us, time allowed for full rotation is 360/6 * 9 * 120 = 64.8ms Fdmax <= 23Hz. • This could be considered the approximate limit for zero-order hold interpolation. • To get higher Doppler tolerance, sophisticated 2-dimensional interpolation is required over the time and frequency, preferably using Weiner filtering. Slide 7
08/07/2010 Conclusions • For delay spreads exceeding 3.5 ms, more than one pilot tone set must be incorporated in the estimation to avoid under-sampling. • When all 3 sets of pilot tones are used, the upper limit of delay spread tolerance is approximately 9.6 ms. • Simple zero-order interpolation in the time domain will not suffice in the presence of Doppler reflectors moving at typical automotive speeds. • A 2-dimensional interpolator using Weiner estimation in both time and frequency dimensions has the potential to yield sufficiently robust channel estimation to both Doppler and Delay spread effects. • Detailed simulations are recommended to evaluate the performance limitations of such interpolation schemes beyond the first-order limitations shown here. Slide 8
Thank you 08/07/2010 Slide 9