LITMIR.BIZ

Скачать книгу

OFDM symbol [2]. The diversity encoder is configured to spread the constellation points from the spatial streams into a plurality of space–time streams in order to provide diversity gain.

      In Figure 2.2, the diversity encoder is shown mapping two spatial streams into four space–time streams (the Number of Spatial Streams N_SS is equal to 2 and the Number of Space–Time Streams [STS] N_STS is equal to 4). Each space–time‐stream corresponds to a different transmitting antenna or a different beam of a beamformed antenna array. The diversity encoder spreads each input constellation point output by the mappers onto first and second output constellation points. The first output constellation point is included in a first space–time stream and the second output constellation point is included in a second space–time stream, different from the first space–time stream. The first output constellation point has a value corresponding to a value of the input constellation point, and the second output constellation point has a value corresponding to a complex conjugate of the value of the input constellation point or to a negative of the complex conjugate (i.e. a negative complex conjugate) [2]. The first output constellation point is at a different time slot (that is, in a different OFDM symbol period) than the second output constellation point when Space–Time Block Coding (STBC) is used. The first output constellation point is at a different frequency (that is, transmitted using a different subcarrier) than the second output constellation point when Space‐Frequency Block Coding (SFBC) is used.

The spatial mapper maps the space–time streams to one or more transmit chains. The spatial mapper maps the space–time stream to the transmit chains using a one‐to‐one correspondence when direct mapping is used. The spatial mapper maps each constellation point in each space–time stream to a plurality of transmit chains when spatial expansion or beamforming is used. Mapping the space–time streams to the transmit chains may include multiplying constellation points of the space–time streams associated with an OFDM subcarrier by a spatial mapping matrix associated with the OFDM subcarrier [2].

      The first to fourth iFTs convert blocks of constellation points output by the spatial mapper to a time domain block (i.e. a symbol) by applying an Inverse Discrete Fourier Transform (iDFT) or an Inverse Fast Fourier Transform (iFFT) to each block. The number of constellation points in each block corresponds to the number of subcarriers in each symbol. A temporal length of the symbol corresponds to an inverse of the subcarrier spacing. When MIMO or MU‐MIMO transmission is used, the TxSP may insert cyclic shift diversities to prevent unintentional beamforming; the cyclic shift diversity may be specified per transmit chain or per space–time stream [2].

      The first to fourth GI inserters prepends a guard interval to the symbol. The TxSP may optionally perform windowing to smooth the edges of each symbol after inserting the GI.

2.5 KEY IEEE 802.11AC MECHANISMS

      New mechanisms were introduced with 802.11ac [18] to increase nominal speed and throughput. A DL channel refers to a communication channel from a transmit antenna of the AP to a receive antenna of a WN/STA, and an UL channel refers to a communication channel from a transmit antenna of a WN/STA to a receive antenna of the AP; DL and UL may be referred to as forward link and reverse link, respectively. New mechanisms 802.11ac included but were not limited to: (i) extended channel binding; (ii) optional 160 MHz and mandatory 80 MHz channel bandwidth for stations; (iii) (as noted), more MIMO spatial streams, also with Downlink Multi‐User MIMO (DL‐MU‐MIMO) – this DL‐MU‐MIMO formulation allows up to four simultaneous clients; (iv) multiple STAs (WNs) each having one or more antennas, to transmit or receive independent data streams simultaneously; (v) 256‐QAM, rate 3/4 and 5/6, added as optional modes (as compared with 64‐QAM, rate 5/6 maximum in 802.11n); and (vi) beamforming with standardized sounding and feedback for compatibility between vendors. Some features (e.g. low‐density parity‐check code; 400 ns short guard interval; five to eight spatial streams; 160 MHz channel bandwidths – contiguous 80 + 80; and 80 + 80 MHz channel bonding including discontiguous sections [19]) are optional.

      2.5.1 Downlink Multi‐User MIMO (DL‐MU‐MIMO)

As discussed earlier, MIMO systems may use multiple transmit antennas to provide beamforming‐based signal transmission. Typically, beamforming‐based signals transmitted from different antennas are adjusted in‐phase (and optionally amplitude) such that the resulting signal power is focused toward a receiver device. See Figure 2.7. A wireless MIMO system may support communication for a single user at a time or several users concurrently; transmissions to a single user (e.g. a single receiver device) are referred to as Single‐User MIMO (SU‐MIMO), while concurrent transmissions to multiple users are referred to as MU‐MIMO. An AP (e.g. a base station [BS]) of an 802.11‐based MIMO system employs multiple antennas for data transmission and reception; each user STA employs one or more antennas. MIMO channels corresponding to transmissions from a set of transmit antennas to a receive antenna are referred to as spatial streams since precoding (e.g. beamforming) is employed to direct the transmissions toward the receive antenna [20]. A MIMO‐based system provides improved performance (e.g. higher throughput and/or greater reliability) using the additional spatial streams.

FIGURE 2.7 Distributed MIMO communication with beamforming [20].

The 802.11n standard introduced MIMO to the LAN environment, allowing a maximum of four MIMO streams to be transmitted to a WN at a time; 802.11ac increased the maximum (theoretical) number of single‐user MIMO streams received by a WN to eight, effectively doubling the network throughput with 802.11ac compared to 802.11n (note that 802.11ac MU‐MIMO specification defines radio configurations that support up to four simultaneous MIMO channels⁵).

Specifically, 802.11ac supports MU‐MIMO affording a major improvement over SU‐MIMO (also just called MIMO). See Figure 2.8. This capability is supported in the DL, and the process is known more specifically as DL‐MU‐MIMO. APs typically have four antennas (APs with eight antennas are also available), but most of the client devices are limited to 1–2 antennas; thus, in a SU‐MIMO channel operation, and the full capacity is rarely achieved. For example, a 4 × 4 Wi‐Fi 11ac AP supports a peak PHY rate of 1.7 Gbps. But a smartphone or tablet with one antenna can only support a peak rate of 433 Mbps, leaving 1.3 Gbps capacity of the AP unused – this difference is called the MIMO gap. 802.11ac addresses the MU‐MIMO gap, allowing an AP to support up to four simultaneous full‐rate Wi‐Fi connections (say, 433 Mbps each) where each of these connections is assigned to a different client device such as smartphone, laptop, or tablet. The total bandwidth of 1.7 Gbps is utilized, representing the systems' bandwidth. In this manner, MU‐MIMO improves performance by affording the AP more options to support the BSS clients and enabling the AP to make full use of the total system throughput. In summary, MU‐MIMO provides increased throughput and reduced latency: the efficient use of available spectrum increases the total capacity of a network by a factor of 2×–3×, and since client devices do not time‐share connections with other clients on the network, each device incurs reduced wait time. To achieve the full MU‐MIMO benefit with an 8 × 8 AP would require an 8 × 8 client configuration; unfortunately, this is not practical, especially with mobile devices and limited battery power (typical mobile clients support 1 × 1 or 2 × 2 configuration). The maximum throughputs in the 5 GHz band are:

FIGURE 2.8 SU‐MIMO versus MU‐MIMO.

Скачать книгу