Microwave Engineering Europe - January/February 2008 - (Page 28)

28 FAST SMALL ANTENNA TEST Figure 3: ”Apparent” and ”effective” diversity gain for two parallel dipoles at the distances 11 mm and 50 mm at 900 MHz. The dashed red line corresponds to the fading of a single dipole normalized to 100 percent efﬁciency. The distance between the dashed red line and the line for selection combining is what we call ”effective” diversity gain, here shown with an arrow. The distance between one of the antenna branches and the line for selection combining is the ”apparent” diversity gain. the same way as described above. After the call has been set-up, the base station simulator sends a bit stream at a given low power to the mobile phone and orders the mobile phone after it has received the bit stream to send it back at full power when it assumes that no further bit errors will occur on the up link. The base station simulator compares the received bit stream with the sent one. If the bit error rate is less than 2.4 percent (example for GSM) the power from the base station simulator is successively lowered in steps until the power that corresponds to a bit error rate of 2.4 percent is found. This power minus the losses in the chamber is the received power at a bit error rate of 2.4 percent. This procedure is repeated for each mode stirrer position. By averaging all the measurements it is possible to calculate the TIS value. TIS measurements shall by deﬁnition be made in a non fading or static environment. This is probably due to the tradition of using anechoic chambers. It is also possible to use a static measurement environment in the Bluetest chambers. The bit error rate is then measured when all mode stirrer mechanisms are in a ﬁxed position. Thismeans that a TIS measurement also takes a long time in a reverberation chamber. The reverberation chamber also offers the possibility of measuring receiver sensitivity in a continuous fading environment, i.e. something which is similar to a real environment. This method is called AFS (Average Fading Sensitivity). The method is similar to what is described above except that it is the average bit error rate that is measured while all mode stirring mechanisms are moving when there is a given power from the base station simulator, i.e. the receiver experiences an environment with Rayeigh fading at a certain average value determined from the base station simulator. It has been shown in several tests done by Bluetest that there is a constant value between AFS and TIS, i.e. TIS can be estimated from AFS. AFS can be measured in about ﬁve minutes which is a considerable improvement in the time to measure receiver sensitivity. If one only wants to measure relative receiver sensitivity of an antenna conﬁguration this can be done in only one minute. Diversity gain Diversity is a technique based on the use of having more than one antenna that experience different fading. By choosing or combining the signals of the different antennas it is possible to gain up to 10-dB, diversity gain, in the worst fading dips at typically 1 percent of the time. It is possible to measure diversity gain with drive tests, i.e. driving or walking with your multiple antenna conﬁguration through a fading environment. The problem when you want to optimize the antenna conﬁguration is that the fading in real environments is always changing and it is not possible to know if the results you get depend on changes in the environment or changes in the antenna conﬁguration. It is also possible to measure diversity gain by measuring each antenna in the antenna conﬁguration separately in an anechoic chamber. After the measurements one can use software to add any type of fading and then estimate the diversity gain. This takes a relatively long time, in the order of one hour. A very efﬁcient alternative is to use the repeatable Rayleigh distribution that is available in the reverberation chamber. The antenna conﬁguration is positioned in the reverberation chamber in the same way as described before. If possible a multi port network analyzer is used to measure amplitude and phase of the different antennas in the antenna conﬁguration and the three ﬁxed antennas in the reverberation chamber S1j. For a diversity antenna with two branches both S12 and S13 are measured simultaneously. Both these antenna branches will show a certain probability to have fading below a certain level, usually referred to as the cumulative distribution function (CDF). By choosing the best of the measured S12 and S13 at every point in time one gets a CDF that is called selection combining. By taking the difference between the CDF for either S12 or S13 and the CDF for selection combining, it is possible to estimate the ”apparent” diversity gain, i.e. how much it is possible to gain in the deepest fading dips, normally at a probability level of 1 percent, by choosing the best antenna. The most relevant value though should be how much you gain in comparison to using an ideal antenna, i.e. to compare the CDF for a single antenna with 100 percent antenna efﬁciency with the CDF for selection combining. This is the ”effective” diversity gain. If you compare with the CDF for a real antenna, i.e. with losses, this is the ”actual” diversity gain. For antenna conﬁgurations with large mutual coupling, as for example two dipoles that are moved very close to each other, their antenna efﬁciency will become very low which means that what may look like a very good diversity gain, i.e. ”apparent” diversity gain, is in reality compared to using just a single antenna — not at all very good. In Figure 3 one can see that at 11-mm distance between two 900-MHz dipoles the ”effective” diversity gain is only 1.5 dB at the 1 percent probability level. For most of the time, in the example in more than 90 percent of the time one will lose signal strength compared to just using a single antenna. In the case of using dipoles the ”effective” and ”actual” diversity gains are very similar since a dipole typically has an antenna efﬁciency of about 95 percent. The measurements depicted in Figure 3 take only one minute to perform in the High Performance chamber. Besides the ”apparent” and ”effective” diversity gain one also obtains the radiation efﬁciency of each antennas, as well as their correlation. MIMO antennas MIMO (Multiple Input Multiple Output) is a technology with multiple antennas at the transmitter and receiver to increase the throughput in new and future wireless communications systems. A necessary condition for multiple antennas to provide higher throughput is that the antennas are relatively uncorrelated, the more the better. The Microwave Engineering Europe ● January/February 2008 ● www.mwee.com 024_025_026_028_029_MWEE.indd 28 24/01/08 15:16:18 http://www.mwee.com

Table of Contents Feed for the Digital Edition of Microwave Engineering Europe - January/February 2008

Microwave Engineering Europe - January/February 2008 Contents News Comment Radio: Raising the Bar for the Radio: Making 802.11n Work Cover Feature: The RF-System-In-Package Trend - Efficient Design with Advanced Design System 2008 Wireless Sensor Networks: The Zigbee PRO Feature Set: More of a Good Thing Very Fast Measurements of Wireless Devices with Small Antennas in Reverberation Chambers WiMAX Update 2008 Bridging the Gap from the CMOS DSP to the Antenna in OFDM Systems Products Calendar

Microwave Engineering Europe - January/February 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.