Archive

Posts Tagged ‘beam forming’

Bang for the buck with explicit beam forming in 802.11ac

October 16th, 2013

 

Bang for the buck with explicit beam forming in 802.11ac

802.11ac has brought with it MIMO alphabet soup … spatial streams, space-time streams, explicit beam forming, CSD, MU-MIMO. Alphabet soup triggers questions to which curious mind seeks answers. This post is an attempt to explore some questions surrounding explicit beam forming (E-BF) that is available in Wave-1 of 802.11ac. E-BF is a mechanism to manipulate transmissions on multiple antennas to facilitate SNR boosting at the target client.

How is E-BF related to spatial streams?

E-BF is a technique different from spatial streams. E-BF can be used whenever there are multiple antennas on the transmitter, irrespective of the number of spatial streams used for transmission.

In the transmission using multiple spatial streams, distinct data streams are modulated on signals transmitted from distinct antennas. The signals from different antennas get mixed up in the wireless medium after they are transmitted from the antennas. The receiver uses signal processing techniques to segregate distinct data streams from the mixture. The ability to separate these distinct data streams from the mixture is dependent on the channel conditions between the transmitter and the receiver (to be able to isolate `S’ streams at the receiver, the channel matrix needs to have rank of `S’ or more). There is no channel dependent processing of signal at the transmitter. Receiver performance is channel dependent. Some key points regarding multiple spatial streams (spatial multiplexing) are:

  • To support `S’-stream transmission, both AP and client must have at least`S’ antennas
  • Rich scattering environment (e.g., indoor) is conducive to give high ranked channel matrix
  • There is no need to send channel feedback from the receiver to the transmitter.

In the transmission using E-BF, the spatial streams are pre-processed (and pre-mixed) to match them to the channel characteristics from the transmitter to the receiver and the output of the pre-processor is transmitted from different antennas. Feedback from the receiver about the  channel characteristics is used in pre-processing. For practical implementation called ZF (Zero Forcing) receiver, E-BF causes SNR boosting for the spatial streams at the receiver. Some key points regarding E-BF are:

  • Feasible with multiple antennas on the AP, irrespective of the number of spatial streams
  • Affects SNR of spatial streams at the receiver
  • Requires channel dependent pre-processing of signal at the transmitter
  • Requires feedback on channel characteristics from the receiver to the transmitter
  • Does not require multiple antennas at the receiver.
When is E-BF truly beneficial?

In general, E-BF is truly beneficial when the number of spatial streams in use is less than the number of antennas on the AP. In Wi-Fi, this most commonly happens when the client has less number of antennas than the AP. For example, most smartphones and tablets have only 1 antenna on them.

Stream vs beam tradeoff:

For the example of 3-stream AP and 3-stream client, adding E-BF on top of 3-stream transmission may not give significant benefit. This is because, with E-BF different spatial streams typically experience unequal boost in SNR. The SNR can be significantly boosted for some spatial streams with E-BF. On the flip side, there will usually also be some spatial streams for which the SNR boost is not significant. Or, it could even be degradation of SNR for some spatial streams compared to the case when E-BF is not used. To be precise, the SNR boost for each spatial stream is dictated by the corresponding singular value of the channel matrix and the singular values of the channel matrix range from high to low for practical channels (E-BF is based on technique called Singular Value Decomposition or SVD). Couple this with the fact that practical implementations use the same MCS on all spatial streams. So, this means either using the MCS supportable by the weakest SNR spatial stream for all spatial streams or using high MCS for the strong streams and dropping the weak streams. There is excellent explanation of this tradeoff  in the book by Perahia and Stacey in Chapter 13 (if you are up for reading some math!).

However, if 3-stream AP can support only 1-stream transmission to the client, E-BF can give significant gain. This is commonly the case with smart devices which have only 1 antenna on them. In this case, the single stream will most likely get boosted in SNR and there isn’t another stream to counter the SNR boost.

How much overhead does E-BF feedback cause on wireless bandwidth?

E-BF requires feedback from the receiver to the transmitter about the channel characteristics. In order to trigger this feedback, the transmitter sends sounding packet to the client. Client performs channel measurements on the sounding packet and responds to the AP with the channel feedback. A question that often comes up in E-BF is how much of a wireless link overhead the E-BF feedback report would cause. To answer that question, take a look at this spreadsheet. From this spreadsheet, it appears that the feedback overhead is relatively small (only about 0.1% of airtime), particularly for the case where E-BF is going to be most beneficial, e.g., 3 or 4-antenna AP talking to 1-antenna client.

All factors considered, E-BF appears to provide benefits for smartphones and tablets, which typically have only 1 antenna and hence cannot support multiple spatial streams when connected to the 3 or 4-stream AP. On the other hand, when there are multiple antennas on both sides of the link (such as a 3 or 4-stream laptop connected to the 3 or 4-stream AP), spatial multiplexing without E-BF can be as good as one with E-BF. These are the inferences drawn from MIMO principles and it would be interesting to see if they match up with measurements from the practical Wave-1 environments.

Earlier Posts in 802.11 Network Engineering Series:

 

802.11ac, 802.11n, WiFi Access , , ,

MU-MIMO: How may the path look like from standardization to implementation?

September 26th, 2013

In earlier blog posts on 802.11ac practical considerations, we reviewed 80 MHz channels, 256 QAM and 5 GHz migration. Continuing the 802.11ac insights series, in this post we will look at some practical aspects of MU-MIMO, which is the star attraction of the impending Wave-2 of 802.11ac.

 

MU-MIMO mechanics and 802.11ac standard

 

Illustration of 802.11ac MU-MIMO

Illustration of 802.11ac MU-MIMO

At a high level, MU-MIMO allows AP with multiple antennas to concurrently transmit frames to multiple clients, when each of the multiple clients has lesser antennas than AP. For example, AP with 4 antennas can use 2-stream transmission to a client which has 2 antennas and 1-stream transmission to a client which has 1 antenna, simultaneously. Implicit requirement to attain such concurrent transmission is beamforming, which has to ensure that bits of the first client coherently combine at its location, while bits of the second client do the same at the second client location. It is also important to ensure that bits of the first client form null beam at the location of the second client and vice versa.

 

What does 802.11ac standard offer for implementing MU-MIMO

  •  The standard provides Group ID Management procedure to form client groups. Clients in a given group can be considered together for co-scheduling of transmissions using the MU-MIMO beamforming.
  • To be able to perform peak/null adjustments in MU-MIMO beamforming as described above, the AP needs to have knowledge of Tx to Rx antennas channel matrix to each client in the group. For this, the standard provides well defined process for channel learning wherein AP transmits sounding packet called as NDP (Null Data Packet) to which clients respond with channel feedback frames (this is called explicit feedback mechanism).

 

 What the standard does not specify

 

There is more to MU-MIMO implementation that is outside of the scope of the standard. The true promise of MU-MIMO is also dependent on these additional implementation factors:

  •  AP has to identify clients that can be co-scheduled in a group. How to form these groups is implementation specific. It is dependent on prevalent channel conditions to different clients. AP will have to make smart decisions on group formation.
  • AP has to keep track of channel conditions for clients in different groups by sending regular sounding packets and receiving explicit feedback to the sounding packets from the clients.  Various implementations may differ based on how frequent channel learning is required in them. Frequent learning increases channel overhead, but may result into cleaner (non-interfering) MU-MIMO beams. Slow learning can result in stale information thereby causing inter-beam interference during concurrent transmissions.
  • When channel conditions change, re-grouping of clients is required. Implementations can differ based on re-grouping triggers and method of re-grouping.
  • Implementations can also differ based on how total antennas on AP are used for beamforming within any given group.
  • The performance of MU-MIMO also depends to some degree on the client side implementation. For demodulating the MU-MIMO signal, clients can implement additional techniques such as interference cancellation to eliminate inter-beam interference.
  • The formation of MU-MIMO groups at physical/MAC layer has to also coincide with traffic and QoS requirements of the clients at higher protocol level.

Practical impact

Practical implementation aspects of MU-MIMOThe above considerations are at practical implementation level. Many of them are in the domain of chip design. How well different chip vendors address them could differentiate them from one another in the MU-MIMO era.

They can also impact Wi-Fi chip design paradigm, which traditionally used similar designs for AP and client radios. With MU-MIMO, there will be bulk of tasks that will be performed at AP, resulting in significant design differences between AP side chipset and client side chipset.

Due to all the nuances of implementation and sensitivity to channel conditions, comparing different MU-MIMO implementations in practical network is difficult task. Notwithstanding, I can imagine MU-MIMO becoming table stake in RFPs after Wave-2 arrives, to which everyone will answer “yes” without heed to the exact implementation details. :-)

One radical thought

Given the cost and complexity of chip level tasks required in MU-MIMO, could there be some chip family which may just use all antennas on the AP to form beam to single client at a time. That is, sequential SU-MIMO, instead of parallel MU-MIMO. What will be pros and cons? Will MU-MIMO be only incrementally or significantly better than sequential SU-MIMO? Time will tell.

Devil is in Detail!

 

Addition Information:

 

802.11ac, Best practices, WiFi Access, WLAN networks , ,