Browse Prior Art Database

ALGORITHM FOR WLAN DEVICES TO EFFICIENTLY DETECT RADAR SYSTEMS BASED ON TEMPORAL AND SPECTRAL CHARACTERISTICS

IP.com Disclosure Number: IPCOM000126790D
Original Publication Date: 2005-Aug-01
Included in the Prior Art Database: 2005-Aug-01
Document File: 5 page(s) / 79K

Publishing Venue

Motorola

Related People

David Vigier: AUTHOR [+3]

Abstract

The 2003 World Radiocommunications Conference agreement that allocated the 5 GHz band to WLAN systems was reached on the basis that WLAN systems would be required to perform Radar Detection in the frequency ranges 5250-5350 MHz and 5470-5725 MHz and evacuate from channels occupied by Radar systems, as described in ITU-R Recommendation M.1652 [1].

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Algorithm for WLAN devices to efficiently detect Radar systems based on temporal and spectral characteristics

David Vigier, Sebastien Simoens, Prakash Moorut

I.                   Introduction

The 2003 World Radiocommunications Conference agreement that allocated the 5 GHz band to WLAN systems was reached on the basis that WLAN systems would be required to perform Radar Detection in the frequency ranges 5250-5350 MHz and 5470-5725 MHz and evacuate from channels occupied by Radar systems, as described in ITU-R Recommendation M.1652 [1].

Recommendation M.1652 mentions that radar pulses are shorter than 20 microseconds. Linked to the fact that WLAN systems that would operate in the 5 GHz band transmit packets longer than 20 microseconds, Radar Detection in its most simple form may be performed by simply differentiating pulse lengths: pulses shorter than 20 microseconds are declared as being Radar pulses, while pulses longer than 20 microseconds are not Radar pulses.

However, military in the US and Europe have recently introduced radar test signals with pulse width longer than 20 microseconds, rendering the previous radar detection schemes based on temporal characteristics useless for such radars.

II.                Description

Such long radar test signals can however be seen as relatively narrowband by a WLAN device which has a 20 MHz bandwidth, taking into account the fact that the WLAN system is based on OFDM and thus uses a Fast Fourier Transform (FFT) block in reception, and that the FFT is calculated every 4 ms using a 3.2 ms window of samples:

·               In

Europe

, fastest chirp rate = 0.8 MHz / 3.2 ms FFT window [2]

·               In the

US

, fastest chirp rate = 1.28 MHz / 3.2 ms FFT window [3]

The idea is therefore to differentiate long Radar pulses based on their bandwidth being lower than the bandwidth of WLAN systems operating in the 5 GHz range. Short Radar pulses are still differentiated from WLAN packets based on their pulse length, as is currently done.

III.             Radar Detection algorithm

The idea described above is derived in an algorithm that is shown in Figure 1.

·               Each signal received is first compared with the Radar Detection threshold as defined by regulation [1]: -62 dBm for devices with maximum EIRP (Effective Isotropic Radiated Power = Tx Power + Antenna Gain) lower than 200 mW, -64 dBm for devices with maximum EIRP higher than 200 mW.

·               If the signal received is above the Radar Detection threshold, then the device tries to synchronize with an OFDM preamble.

·               If no OFDM preamble is detected, then the receiver estimates the signal length, as well as the signal bandwidth.

·               If the pulse length is below the minimum WLAN packet duration, then the pulse is declared as being a potential radar pulse. A WLAN frame is necessarily composed of a PLCP (Physical Layer Convergence Procedure) preamble which is 16 microseconds long, a Signal field which is 4 microseconds long, and a Data field at least 4 microseconds long (the duration of one OFDM symbol, including th...