org.kurento.client

Class RembParams

java.lang.Object

org.kurento.client.RembParams

public class RembParams
extends Object

Defines values for parameters of congestion control

Constructor Summary

Constructors

Constructor and Description
RembParams() Create a RembParams

Method Summary

Modifier and Type	Method and Description
Float	getDecrementFactor() get Determine how much is decreased the current REMB when too losses are detected.
Float	getExponentialFactor() get Factor used to increase exponentially the next REMB when it is below the threshold.
Float	getLinealFactorGrade() get Determine the value of the next linearFactor based on the threshold and the current REMB.
Integer	getLinealFactorMin() get Set the min of the factor used to increase linearly the next REMB when it is over the threshold.
Integer	getPacketsRecvIntervalTop() get Size of the RTP packets history to smooth fraction-lost.
Integer	getRembOnConnect() get Initial local REMB bandwidth estimation that gets propagated when a new endpoint is connected.
Float	getThresholdFactor() get Determine the next threshold (TH) when too losses are detected.
Integer	getUpLosses() get Max fraction-lost to no determine too losses.
void	setDecrementFactor(Float decrementFactor) set Determine how much is decreased the current REMB when too losses are detected.
void	setExponentialFactor(Float exponentialFactor) set Factor used to increase exponentially the next REMB when it is below the threshold.
void	setLinealFactorGrade(Float linealFactorGrade) set Determine the value of the next linearFactor based on the threshold and the current REMB.
void	setLinealFactorMin(Integer linealFactorMin) set Set the min of the factor used to increase linearly the next REMB when it is over the threshold.
void	setPacketsRecvIntervalTop(Integer packetsRecvIntervalTop) set Size of the RTP packets history to smooth fraction-lost.
void	setRembOnConnect(Integer rembOnConnect) set Initial local REMB bandwidth estimation that gets propagated when a new endpoint is connected.
void	setThresholdFactor(Float thresholdFactor) set Determine the next threshold (TH) when too losses are detected.
void	setUpLosses(Integer upLosses) set Max fraction-lost to no determine too losses.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

RembParams

public RembParams()

Create a RembParams

Method Detail

getPacketsRecvIntervalTop

public Integer getPacketsRecvIntervalTop()

get Size of the RTP packets history to smooth fraction-lost. Units: num of packets

setPacketsRecvIntervalTop

public void setPacketsRecvIntervalTop(Integer packetsRecvIntervalTop)

set Size of the RTP packets history to smooth fraction-lost. Units: num of packets

getExponentialFactor

public Float getExponentialFactor()

get Factor used to increase exponentially the next REMB when it is below the threshold. REMB[i+1] = REMB[i] * (1 + exponentialFactor)

setExponentialFactor

public void setExponentialFactor(Float exponentialFactor)

set Factor used to increase exponentially the next REMB when it is below the threshold. REMB[i+1] = REMB[i] * (1 + exponentialFactor)

getLinealFactorMin

public Integer getLinealFactorMin()

get Set the min of the factor used to increase linearly the next REMB when it is over the threshold. Units: bps (bits per second). REMB[i+1] = REMB[i] + MIN (linealFactorMin, linealFactor)

setLinealFactorMin

public void setLinealFactorMin(Integer linealFactorMin)

set Set the min of the factor used to increase linearly the next REMB when it is over the threshold. Units: bps (bits per second). REMB[i+1] = REMB[i] + MIN (linealFactorMin, linealFactor)

getLinealFactorGrade

public Float getLinealFactorGrade()

get Determine the value of the next linearFactor based on the threshold and the current REMB. Taking into account that the frequency of updating is 500ms, the default value makes that the last REMB is reached in 60secs. linealFactor = (REMB - TH) / linealFactorGrade

setLinealFactorGrade

public void setLinealFactorGrade(Float linealFactorGrade)

set Determine the value of the next linearFactor based on the threshold and the current REMB. Taking into account that the frequency of updating is 500ms, the default value makes that the last REMB is reached in 60secs. linealFactor = (REMB - TH) / linealFactorGrade

getDecrementFactor

public Float getDecrementFactor()

get Determine how much is decreased the current REMB when too losses are detected. REMB[i+1] = REMB[i] * decrementFactor

setDecrementFactor

public void setDecrementFactor(Float decrementFactor)

set Determine how much is decreased the current REMB when too losses are detected. REMB[i+1] = REMB[i] * decrementFactor

getThresholdFactor

public Float getThresholdFactor()

get Determine the next threshold (TH) when too losses are detected. TH[i+1] = REMB[i] * thresholdFactor

setThresholdFactor

public void setThresholdFactor(Float thresholdFactor)

set Determine the next threshold (TH) when too losses are detected. TH[i+1] = REMB[i] * thresholdFactor

getUpLosses

public Integer getUpLosses()

get Max fraction-lost to no determine too losses. This value is the denominator of the fraction N/256, so the default value is about 4% of losses (12/256)

setUpLosses

public void setUpLosses(Integer upLosses)

set Max fraction-lost to no determine too losses. This value is the denominator of the fraction N/256, so the default value is about 4% of losses (12/256)

getRembOnConnect

public Integer getRembOnConnect()

get Initial local REMB bandwidth estimation that gets propagated when a new endpoint is connected.

The REMB congestion control algorithm works by gradually increasing the output video bitrate, until the available bandwidth is fully used or the maximum send bitrate has been reached. This is a slow, progressive change, which starts at 300 kbps by default. You can change the default starting point of REMB estimations, by setting this parameter.

WARNING: If you set this parameter to a high value that is higher than the network capacity, then all endpoints will start already in a congested state, providing very bad video quality until the congestion control algorithm is able to recover from the situation. Network congestion is very unpredictable, so be careful when changing this parameter; for most use cases it is safer to just start with a low initial value and allow the REMB algorithm to raise until the optimum bitrate is reached.

• Unit: bps (bits per second).
• Default: 300000 (300 kbps).

setRembOnConnect

public void setRembOnConnect(Integer rembOnConnect)

set Initial local REMB bandwidth estimation that gets propagated when a new endpoint is connected.

The REMB congestion control algorithm works by gradually increasing the output video bitrate, until the available bandwidth is fully used or the maximum send bitrate has been reached. This is a slow, progressive change, which starts at 300 kbps by default. You can change the default starting point of REMB estimations, by setting this parameter.

WARNING: If you set this parameter to a high value that is higher than the network capacity, then all endpoints will start already in a congested state, providing very bad video quality until the congestion control algorithm is able to recover from the situation. Network congestion is very unpredictable, so be careful when changing this parameter; for most use cases it is safer to just start with a low initial value and allow the REMB algorithm to raise until the optimum bitrate is reached.

• Unit: bps (bits per second).
• Default: 300000 (300 kbps).