WO2004112283A1

WO2004112283A1 - Transmit power control outer loop for improving speech recognition rate management in mobile communications systems

Info

Publication number: WO2004112283A1
Application number: PCT/KR2003/001174
Authority: WO
Inventors: Dong Woo Kim
Original assignee: Kwangwoon Foundation
Priority date: 2003-06-16
Filing date: 2003-06-16
Publication date: 2004-12-23
Also published as: AU2003234941A1

Abstract

Disclosed is a transmit power control outer loop for improving a speech recognition success rate in a mobile communication system, which adaptively controls a target Frame Error Rate as a target power control value corresponding to a Speech Recognition Success Rate by adding a Speech Recognition Rate Management outer loop to a power control loop wherein the power control loop consists of an inner loop and an outer loop. In order to achieve that, in the transmit power control loop for improving speech recognition success rate, a Speech Recognition Rate Management outer loop having a Speech Recognition Success Rate estimator is included to an outer loop. Speech Recognition Success Rate estimator estimates the rate of successful recognition from received voice frames. Speech Recognition Rate Management outer loop compares estimated Speech Recognition Success Rate with target Speech Recognition Success Rate and automatically adjusts the target Frame Error Rate according to the compared result.

Description

TRANSMIT POWER CONTROL OUTER LOOP FOR IMPROVING SPEECH RECOGNITION RATE MANAGEMENT IN MOBILE COMMUNICATIONS

SYSTEMS

Technical Field

The present invention relates to a transmit power control outer loop for improving a speech recognition rate in a mobile communications system. More specifically, the present invention relates to a transmit power control outer loop for improving a speech recognition rate in a mobile communication system, which adaptively control a target Frame Error Rate as a target power control value corresponding to a target Speech Recognition Rate by adding a Speech Recognition Rate Management outer loop to a power control loop wherein the power loop consists of an inner loop and an outer loop.

Background Art

Generally, in order to provide a Speech Recognition Service (SRS) through a wireless channel of a mobile communication system, the SRS requires transmission quality which is better than that of a conventional Ordinary Voice Service (OVS). Transmission quality in the mobile communication system changes according to the state of a wireless channel. Accordingly, the present invention uses a transmit power control loop which adjusts a target transmission quality value according to a change in a wireless channel state in order to obtain a target Speech Recognition Rate.

Due to the great demand for a mobile communication service, the technology of the mobile communication system has been developed rapidly. In addition, various kinds of applied services have been provided through a wireless channel of the mobile communication system.

The SRS is one of the applied services provided by the mobile communication system. For example, the SRS is used for speech recognition dialing, speech recognition based stock dealings, or speech recognition based web browsing. The mobile communication system also provides such SRS.

In general, the mobile communication system provides services through a wireless channel. Transmission quality in the wireless channel changes in a moment according to a length, buildings, or natural circumstance between a transmitter and a receiver, a moving speed of the receiver.

In order to compensate that, the mobile communication system uses a transmit power control loop. When the transmit control loop has transmission quality less than a target value, it increases a transmit power. When the transmit control loop has the transmission quality greater than the target value, it reduces the transmit power.

Where the transmission quality is greater than the target value, the transmit power is reduced to prevent an unnecessary consumption of a battery and to reduce an interference signal magnitude with respect to other receivers. Hereinafter, a conventional transmit control loop will be described with reference to FIG. 1.

FIG. 1 is a view showing a transmit power control loop in a mobile communication system which estimates a Signal-to-interference ratio (SIR) as transmission quality on a wireless channel and which sets a Frame Error Rate (FER) as user sensible transmission quality.

A transmit power control is formed by a discrete time unit expressed by a constant t. The conventional power control shown in FIG. 1 includes an inner loop and an outer loop. In the operation of the inner loop, when a transmitter Tx transmits a transmission power P_(/) at time t, a path combiner of a receiver combines received

signals for a plurality of paths between the transmitter and the receiver and measures SIR F^w .

The inner loop compares the measured SIR F⁽⁰ with a target SIR χ^{t) .

When the measured SIR F^w is greater than the target SIR γ^ , the inner loop

transmits a Transmit Power Control (TPC) command indicating an increase in power to the transmitter. When the measured SIR F^w is less than the target SIR γ^w , the inner loop transmits a TPC command indicating a reduction in power to the

transmitter.

In the place of the inner loop, the target SIR γ^{i) is an externally given

value. The outer loop determines the externally given value γ^ .

A Frame Error Detector receives signals for the period of one frame and determines whether, an error occurs in a received frame. The Frame Error Detector transmits the error occurrence result to a Frame Error Counter and changes a Frame Error Rate measured value F^{t) of the Frame Error Counter by a predetermined period T_frame.

The Frame Error Rate measured value F^{t) is compared with a target Frame Error Rate Φ⁽⁰. When F^(t) is greater than Φ^(t) , the outer loop reduces γ^{t) . When

F^{t) is less than Φ⁽⁰ , the outer loop increases γ^{() . When F^{l) is identical with

Φ⁽⁰ , the outer loop uses a previous value as a target Frame Error Rate.

Accordingly, a power P⁽'⁺¹⁾ to be transmitted at time t+1 is determined by P^(t) and a TPC command received as a result of the transmit power control loop.

The target SIR χ^w is expressed by equation 1.

[Equation 1]

Outer loop

_γ(t)

where the Aχ is a step size for adjusting a target SIR, and a value suitable

for a special mobile communication system is previously set as Aγ .

[Equation 2]

Inner loop

JD(H- 1) ;= J P^α) + AP if F^{( i)} < γ^{{ t)} p^ - AP if r^(t) > γ^ω

where AP is a step size for adjusting a transmission power value, and a value suitable for a special mobile communication system is previously set as AP .

As shown in FIG. 1, in the conventional transmit power control method, a threshold for a power control is determined to obtain a target FER Φ⁽⁰ . In a mobile communication system for providing OVS, 1 % is set as Φ^w regardless of time.

This is caused by an experimental result in which suitable call equality of OVS is maintained when an FER performance becomes about 1 %.

However, in a mobile communication system providing an SRS, a value less than SRS is set as the target Frame Error Rate Φ^w and the set target Frame Error Rate Φ^(r) is adjusted according to a Speech Recognition Rate.

Without an adjustment of the target Frame Error Rate Φ^w , obtaining of a target Speech Recognition Rate is impossible in a mobile communication system in which a magnitude of a signal changes.

Disclosure of Invention

Therefore, it is an object of the present invention to provide a transmit power control outer loop for improving a speech recognition rate in a mobile communication system, which adaptively controls a target Frame Error Rate, as a target power control value corresponding to a target Speech Recognition Rate by adding a Speech Recognition Rate Management outer loop to with an outer loop of a power control loop.

According to the present invention, there is provided a transmit power control outer loop for improving a speech recognition rate in a mobile communication system

Brief Description of the Drawings

The foregoing and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which:

FIG. 1 is a view for illustrating a conventional transmit power control loop in a mobile communication system; and

FIG. 2 is a transmit power control outer loop for improving a speech recognition rate in a mobile communication system according to an embodiment of the present invention.

Best Mode for Carrying Out the Invention

Hereinafter, a preferred embodiment of the present invention will be described with reference to FIG. 2.

In a mobile communication environment, when a Frame Error Rate is increased by a radio fading, an outer loop increases a target Signal-to-Interference ratio in order to obtain a target FER. At this time, in a speech recognition applied service, a problem occurs on how much of a target FER should be determined in order to obtain the target SRR. The present invention provides a method, which automatically supports a target SRR by an SRRM outer loop.

A distinct feature of the present invention is to add a Speech Recognition Rate Management to a transmit power control loop of a mobile communication system which controls a transmission power by an inner loop and an outer loop. In detail, in order to embody an SRS of good quality, an SRRM outer loop according to the present invention is included to the outer loop of a transmit power control loop.

A Frame Error Detector of the outer loop detects whether an error occurs in a frame. The SRRM outer loop includes an SRR estimator which receives the frame from the Frame Error Detector.

Accordingly, an SRR estimator according to the present invention is installed at a receiver of the mobile communication system. The SRR estimator measures an SRR from the received frame. The measured SRR is compared with a target SRR. A currently set target FER is automatically changed according to the compared result.

Hereinafter_/ an operation of an SRRM outer loop according to the present invention will be described.

The Speech Recognition Rate estimator estimates the Speech Recognition Rate Ho using the received frame. The estimated Speech Recognition Rate H^w is compared with a target Speech Recognition Rate η^{t) . When the estimated Speech

Recognition Rate H^w is greater than the target Speech Recognition Rate 77 ^w , a

target Frame Error Rate Φ⁽⁰ is reduced. When the estimated Speech Recognition Rate H^w is less than the target Speech Recognition Rate η^ , the target Frame

Error Rate Φ^w is increased. When the estimated Speech Recognition Rate H^{t) is identical with the target Speech Recognition Rate η^{t) , a previous Frame Error Rate

is used for the target Frame Error Rate Φ⁽⁰.

According to the equations 1 and 2, the inner loop receives a power P⁽⁽⁾ and determines P^{M) based on the power P⁽⁰. A target FER is used to determine the P^{M) . The target FER Φ⁽⁰ is determined by the following equation 3 to maintain a target Speech Recognition Rate /7^W in an SRRM loop.

[Equation 3] SRRM outer loop

where ΔΦ is a step size for adjusting a target FER, and a value suitable for a special mobile communication system is previously set as the step size ΔΦ .

The target Speech Recognition Rate 77 ^w mainly has a fixed value, or an

externally variable target value is set as the target Speech Recognition Rate 77 ^w .

According to the present invention, a target Frame Error Rate of a transmit power control loop is automatically set depending on whether a Speech Recognition

Rate becomes a target value. Accordingly, when an SRS is provided through a wireless channel in a mobile communication system, an intensity of a transmission power is determined to obtain a target recognition rate.

Since the present invention adjusts the target Speech Recognition Rate η^{t)

according to a time zone and a kind of an SRS applied service, it is used for various kinds of target SRS recognition rates.

Industrial Applicability

As can be seen from the foregoing, with accordance to a transmit power control loop for improving a speech recognition rate in a mobile communications systems according to the present invention, it is a speech recognition rate management based power control loop in order to provide an SRS through a wireless channel in the mobile communication system. The present invention easily supports a given target recognition rate by automatically adjusting/ controlling a target Frame Error Rate by means of an SRMM loop.

When used codecs or recognition engines are different from each other, the conventional transmit power control loop obtains a different SRR in spite of using the same FER. However, although different codecs or recognition engines are used, the present invention provides the advantageous calculating method, in which FER is calculated to correspond to the moving environment in order to obtain the target SRR.

Claims

1. A transmit power control outer loop including inner and outer loops, the transmit power control outer loop comprising: a Speech Recognition Rate Management outer loop included to the outer loop and having a Speech Recognition Rate estimator for estimating a Speech Recognition Rate from a received frame, the Speech Recognition Rate Management outer loop comparing the Speech Recognition Rate estimated by the Speech Recognition Rate estimator with a target Speech Recognition Rate, and automatically changing set target Frame Error Rate according to the compared result.

2. The transmit power control outer loop according to claim 1, wherein the Speech Recognition Rate Management outer loop estimates the Speech Recognition Rate H^{t) using the received frame by the Speech Recognition Rate estimator; compares the estimated Speech Recognition Rate H⁽⁰ with a target Speech

Recognition Rate η^{t) , reducing a target Frame Error Rate Φ^w when the estimated

Speech Recognition Rate H^w is greater than the target Speech Recognition Rate η^(t) , increasing the target Frame Error Rate Φ^w when the estimated Speech

Recognition Rate H^w is less than the target Speech Recognition Rate η^{t) , and

using a previous Frame Error Rate when the estimated Speech Recognition Rate H^w is identical with the target Speech Recognition Rate η ^{t) .

3. The apparatus according to claim 1, wherein the SRR estimator receives a frame wherein an error occurs in the frame by a Frame Error Detector.