MIMO isn't a new technology and its basic capabilities are used in 802.11n Wi-Fi systems. However, it's a mistake to assume that experience with MIMO in a fixed environment such as Wi-Fi is directly applicable to mobile WiMAX for the reason that mobile WiMAX supports two types of MIMOs. The so-called rate 1 version uses spatial multiplexing to improve the effective signal strength, whereas the rate 2 version uses space-time coding techniques to send two different signals on the same frequency as long as signal strength is high and there are two different channels between the multiple antenna elements at the base station and the terminal. Figure 1 highlights the key principles of spatial multiplexing techniques used in mobile WiMAX MIMO.

In mobile WiMAX, when the signal conditions are weak, space-time coding is used to maintain a downlink throughput level that provides a good user experience. When a strong signal becomes available — such as when the device moves closer to another base station — the network and device should switch to spatial multiplexing, which potentially doubles throughput. The network determines which technique to use based on signal strength and channel estimation reports from the user device. That sounds straightforward, but it's a complex process that, if handled incorrectly, has a negative impact on the network. For example, if the user device frequently miscalculates the channel characteristics, then the base station wastes resources and airlink capacity when switching between space-time coding and spatial multiplexing unnecessarily.

This flaw also affects network capacity. For example, if the user device fails to report that channel characteristics have changed, then the base station is transmitting data at a rate far higher than the channel can support. As a result, many packets may have to be retransmitted because they're lost or corrupt due to the poor channel conditions. Those retransmissions waste sector capacity that could have been provided to other users. In the other extreme of improving channel characteristics, fewer data is sent than the channel can support and the airlink capacity could have been much higher had the base station switched the connection to the spatial multiplexing mode.

This situation highlights why not all MIMO solutions will be of equal performance, and it could be a mistake to assume that a chipset vendor's experience with MIMO in a fixed environment is applicable to mobility. In a mobile environment, where signal conditions change more frequently and rapidly, the algorithms designed for a stable fixed environment — such as in the 802.11n world — can't keep up. To avoid these problems, user devices should have Wave 2 chipsets that accurately calculate and report channel characteristics.

The challenge to properly estimate channel conditions and the sophisticated algorithms needed to implement MIMO is one reason why mobile WiMAX device vendors should choose Wave 2 chipsets that are designed for a rigorous mobile environment, and are proven in years of field trials. For example, Beceem's mobile WiMAX chipsets have been in field trials since October 2005. This experience has helped refine its algorithms to the point that they're sophisticated enough to report channel characteristics with high accuracy, even when users are moving at a fast pace, with frequent hand-offs.

In addition, the company has been able to use its field and development experience to improve the performance of its RFIC products. One example is the performance of the direct-conversion RFIC included in the Wave 2 chipset, the BCSR200. RFICs are generally expected to receive and downconvert the radio signal that is transmitted from the base station with minimum distortion, and pass a highly accurate signal to the baseband chip for signal processing. Only if the baseband chip receives a highly accurate signal from the radio can it fully use the advanced smart antenna features of mobile WiMAX and maximize the performance and spectral efficiency.

One critical proof for the effectiveness of a RFIC is measurements of the signal integrity under the most challenging conditions. The WiMAX OFDMA waveform, with its many subcarriers, is best analyzed after demodulation of the most challenging constellation, 64 QAM. Figure 2 shows the screenshot of a 64 QAM measurement with Agilent VSA software. One can see the relative constellation error (RCE), sometimes referred to as error vector magnitude (EVM), of each subcarrier and a composite EVM of all subcarriers after demodulation and measurement with the VSA. The exceptionally low EVM of -35 dB demonstrates that Beceem's full radio chain has a low noise floor, high compression point, and low composite phase noise. The -35 dB EVM allows the WiMAX modem to receive high data rate 64 QAM symbols with almost no degradation to the data. This level of performance will allow very high data rates with low bit error rates (BER) under high SNR conditions or allow the full margin of SNR reduction to be allocated to the external link budget for moderate SNR conditions.

Previous 1 2 3 4 Next

RSS    Save to Del.icio.us  Digg This