SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 501 | Next

Benny Bing

"Emerging Technologies in Wireless LANs: Theory, Design, and Deployment"

7.7 Examples
Previously, we saw that eigenvector beamforming coupled with a simple MMSE-based
receiver can outperform spatial spreading even when an iterative receiver is employed
together with the latter. In this section, we quantify the computational complexity
associated with these two approaches [12]. Specifically, we consider the 2?—2 system
operating at 39 Mbps introduced in Section 9.6.3. The number of information bits is Nb =
MIMO Spatial Processing for 802.11n WLAN 214
8000. Table 9.4 shows the equivalent number of real MUL and ADD operations for the
following four configurations:
1. Spatial spreading; 2 streams of QPSK, R = 3/4; NO = 52; iterative LSD-SOVA receiver
with Nit = 1 iteration (Nave(1) = 2.20; Nave(2) = 2.33).
2. Same as 1., but with Nit = 2 iterations.
3. Eigenvector beamforming; 1 stream of 16-QAM, R = 3/4; NO = 52; MMSE receiver.
4. Eigenvector beamforming; 2 streams, 16-QAM, QPSK, R = 1/2; NO = 52; MMSE
receiver.
Table 9.4: Complexity comparison of iterative
LSD and MMSE-based receivers.
Real MUL Real ADD
1. 40.12 440.90
2. 45.25 1122.02
3. 19.86 453.66
4. 27.95 465.30
Note that the fourth configuration employs different modulations on the two streams.


Pages:
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513