The problem with ABR is that the client device, such as a smartphone or tablet, is in charge of the bandwidth and isn’t fair about how that capacity is allocated. If an iPhone is the first device on the home network to request a video stream, it will typically receive a high bit-rate version — perhaps more than it really needs. Then, when a connected HD television requests a stream, it tends to get the scraps, resulting in a crummy-looking pixel-icious image.
via Light Reading Cable – IP & Convergence – Ericsson Adapts to the TV Streaming Challenge – Telecom News Analysis.
Ericsson AB (Nasdaq: ERIC) is trying to solve the problem by applying Weighted Fair Queuing (WFQ), a data packet scheduling technique, to ABR streams.
From Wikipedia on WFQ:
WFQ is a generalization of fair queuing (FQ). Both in WFQ and FQ, each data flow has a separate FIFO queue. In FQ, with a link data rate of
, at any given time the
active data flows (the ones with non-empty queues) are serviced simultaneously, each at an average data rate of
. Since each data flow has its own queue, an ill-behaved flow (who has sent larger packets or more packets per second than the others since it became active) will only punish itself and not other sessions.
As opposed to FQ, WFQ allows different sessions to have different service shares. If
data flows currently are active, with weights
data flow number
will achieve an average data rate of
![\frac{Rw_i}{(w_1+w_2+...+w_N)}](http://upload.wikimedia.org/math/4/a/6/4a63205bc3cdcfcc14a1b067cfdc04b1.png)
It can be proven [1] that when using a network with WFQ switches and a data flow that is leaky bucket constrained, an end-to-end delay bound can be guaranteed. By regulating the WFQ weights dynamically, WFQ can be utilized for controlling the quality of service, for example to achieve guaranteed data rate.