WO2001069962A1

WO2001069962A1 - Use of silence detection for improved received voice quality during a search of a neighbor set

Info

Publication number: WO2001069962A1
Application number: PCT/US2001/004013
Authority: WO
Inventors: John Michael Belstner
Original assignee: Ericsson Inc.
Priority date: 2000-03-14
Filing date: 2001-02-08
Publication date: 2001-09-20
Also published as: CN1418445A; AU2001236759A1

Abstract

A method of a mobile terminal searching signal strength of neighboring base stations within a cellular communications network. The method includes receiving a signal, transmitted at a first frequency, at a mobile terminal from a first base station. The received signal is processed at a signal processing unit within the mobile terminal. The received signal is analyzed and a period of silence within the signal is determined. Thereafter, the mobile teminal ceases signal processing and searches alternate frequencies of neighboring base stations and determines a signal strength measurement for each of the neighboring base stations. After performing the neighboring base station searching, the mobile terminal restores signal processing of the signal through the first base station at the first frequency.

Description

USE OF SILENCE DETECTION FOR IMPROVED RECEIVED VOICE QUALITY DURING A SEARCH OF A NEIGHBOR SET

Field of the Invention

The present invention is directed to a method of searching neighboring base stations for determining handoffs of an active call and, more particularly, to a method of performing alternate frequency neighbor searches during periods of silence in the received audio of the active call to reduce or eliminate audio distortion caused by interrupted words or phrases.

Background of the Invention

One of the requirements of a Code Division Multiple Access (CDMA) mobile terminal is to perform the function of mobility management or "set maintenance" as it is referred to within TIA/EIA-95. To meet this requirement, the mobile terminal continually searches a set of neighboring base stations or "pilots" to find one that may be more suitable for operation. This list of neighboring base stations is supplied to the mobile terminal by the base station.

When one of the neighboring pilots to be searched is on a different CDMA frequency, voice processing on the current frequency must cease momentarily while the new frequency is loaded and a pilot strength measurement is taken. The time taken before the previous CDMA frequency is restored and voice processing is allowed to continue causes a "hole" in the received speech. In one embodiment, each frequency requires about 20 milliseconds to search before the voice processing is restored. When there is more than one frequency to search, the "hole" becomes substantial and voice quality greatly decreases as it may be in the middle of a syllable, which could change the understanding of a word or phrase.

Currently, most CDMA networks transmit over a single frequency, and the mobile terminals do not search within alternate frequencies and therefore are not forced to stop audio processing. However, as more networks are installed and the number of subscribers increase, CDMA service providers will have no choice but to utilize more than one frequency.

Time Division Multiple Access (TDMA) networks have a similar dilemma when searching for neighbor base stations. One technique currently being used in TDMA systems to reduce the annoying holes is to stagger the search intervals. This reduces the repetitiveness of the holes and reduces the annoyance somewhat, but does not completely eliminate the problem.

Thus, there remains a need for a method to perform alternate frequency neighbor searches without causing audio distortion to an active call.

Summary of the Invention

The present invention utilizes the naturally occurring pauses during a conversation to perform alternate frequency strength measurements from neighboring base stations. The strength measurements are performed during these pauses resulting in little to no distortion occurring in the received audio that is heard by the user.

The steps of the invention include receiving an incoming signal at a receiver and processing the signal at a digital signal processor. Audio portions of the signal are directed to a vocoder and sent to a speaker to be heard by the user. Outgoing messages spoken by the user are directed to the vocoder and sent to the digital signal processor where they are sent to the base station by a transmitter. During the processing of incoming and outgoing signals, a period of silence is detected at the digital signal processor. During the silence, the signal processing is halted as alternate frequency tests are performed for strength measurements of neighboring base station frequencies for possible hand-off candidates. Once the strength measurements have been completed, the incoming and outgoing signals are again processed through the digital signal processor.

This method provides for seamless audio that is unnoticeable to the user. During the alternate frequency search, previously stored audio may be fed to the vocoder to prevent any breaks, pops, hisses, or other like disturbances from occurring which may be noticeable to the user. Varying search rates may further be used to prevent noticeable distortion.

Brief Description of the Drawings

Figure 1 is a schematic illustration of a cellular network in accordance with the present invention;

Figure 2 is a schematic diagram illustrating the elements of a mobile terminal used within the present invention; and

Figure 3 is a flowchart diagram illustrating the steps in performing an alternate frequency neighbor search within a wireless communication network. Detailed Description of the Invention

In the following description, like reference characters designate like or corresponding parts throughout the several views. Referring now to the drawings, the method of alternate frequency neighbor searching of the present invention is described. This method is useful in mobile communication networks like that illustrated schematically in Figure 1. The mobile communication network, which is indicated generally by the numeral 10, includes a plurality of base stations 12, which are connected via a mobile services switching center (MSC) 14 to a terrestrial communications network such as the Public Switched Telephone Network (PSTN) 18. Each base station 12 is located in and provides service to a geographic region referred to as a cell. In general, there is one base station 12 for each cell within a given mobile communication network 10. Within each cell, there may be a plurality of mobile terminals 16 that communicate via a radio link with the base station 12. The base station 12 allows the user of the mobile terminal 16 to communicate with other mobile terminals 16, or with users connected to the PSTN 18. The MSC 14 routes calls to and from the mobile terminal 16 through the appropriate base station 12.

A handoff occurs when an active call currently being processed by a first base station 12 is switched to and processed by a different base station. One situation requiring a handoff occurs when the received signal quality drops below acceptable levels and another available base station 12 having a higher signal quality is available to process the call. Network management to increase system capacity may also force an active call to be handed-off from a first base station 12, to a second base station 12 having more available capacity. The mobile terminal 16 performs signal strength measurements of the neighboring base stations 12 to determine when a handoff is required in accordance with TIA/EIA-95. The serving base station 12 transmits a list of neighboring base stations, commonly referred to as a neighbor list, at the start of a call or after a handoff. The neighbor list identifies frequencies of neighboring base stations 12 that are potential handoff targets. The mobile terminal 16 continually reports the results of these measurements to the serving base station 12. The mobile terminal 16 may perform signal strength measurements on a regular schedule, or may vary the rate at which the measurements are taken.

The measurement reports provided by the mobile terminal 16 give the serving base station 12 a list of the signal strengths from adjacent cells, as measured by the mobile station 16 at its present location. The mobile communication network 10 also knows which adjacent cells have unused radio channels that are available for allocation during a handoff. From the list of available channels, the mobile communication network 10 selects the cell which best can handle the call from a service quality and an overall interference point of view based on signal strength and bit error rate. A suitable traffic channel assigned the target base station 12 is selected, and the mobile terminal 16 is commanded to tune to the selected traffic channel in the target cell. At the same time, the call is switched by the MSC 14 to the base station 12 currently serving the mobile terminal 16 to the base station 12 in the target cell.

Figure 2 illustrates the mobile terminal 16 used within a CDMA mobile network. The mobile terminal 16 includes a receiver 102 to receive transmitted signals. A digital signal processor (DSP) 104 extracts voice data from the received signal and feeds the information to a vocoder 110 where it is output through speaker 114. For transmitting outgoing messages, audio received from a microphone 112 is converted into digital code at the vocoder 110. The signal is then processed by the DSP 104 where it is expanded using a spread spectrum technique that spreads the voice stream over the full 1.25 MHz of a CDMA channel. The vocoder 110 output is sent to a transmitter 116 which transmits the signal. A central processing unit (CPU) 106 receives control/data messages for controlling the function of the mobile terminal 16. Memory 107 may store received audio for sending to the vocoder 110 during search periods as will be described below.

A CDMA signal received by the receiver 102 is quantized and filtered through a Walsh code correlator which identifies the call from the other signals within the frequency. Additionally, the signal is processed through a psuedo-noise sequence correlator which removes a psuedo-noise (PN) that was added to further extract the desired signal from other signals transmitted or the same frequency.

The recovered signal contains both voice/audio messages and control/data messages. The voice/audio messages are processed through a Viterbi decoder schematically positioned within the DSP 104 that detects and corrects errors introduced in the signal during transmission. Preferably, the DSP 104 decodes the signal after receiving a predetermined number of bits. The incoming signals are stored at the DSP 104 until receiving this predetermined number of bits, or until a predetermined time period has expired without receiving any additional signals. In one embodiment, the DSP 104 waits for about 20 milliseconds of voice/audio data to be received prior to decoding the message. Once the voice/audio message is decoded, it is sent to the vocoder 110 which converts the received bits into analog audio waveforms that are output through the speaker 114.

Outgoing messages are processed in the inverse manner. Audio received by the microphone 112 is passed to the vocoder 110 where it is digitally compressed and converted into digital code. The message is then forwarded to the DSP 104 where it is encoded by a convolutional encoder, and orthogonally coded using a Walsh code and PN sequence. The signal is then sent to the transmitter 116 where it is transmitted to the base station 12 handling the call.

Within both the incoming voice/audio messages, and outgoing messages, there are natural pauses within the data streams. These pauses, or moments of silence, are naturally occurring within human speech such as the time between syllables, words, or sentences. During these pauses, there is no voice/audio data for the DSP 104 to process. These periods of silence provide time to search neighboring frequencies. Nearly all the naturally occurring pauses in human speech are longer than the time needed to perform the search.

For incoming calls, during the time that the DSP 104 is searching, no voice/audio signal is sent to the vocoder 110. The incoming signals are stored in the DSP 104 for processing after the frequency search has been completed. The absence of a signal being sent to the vocoder 110 may result in a buzz or pop that is audible through the speaker 114. Additionally, the absence of a signal may cause a change in the background noise level during the call which is also noticeable to the user. To ensure that the searches go unnoticed by the user, the DSP 104 may send previously stored voice/audio signals stored in the memory 107 as background noise to the vocoder 110. The vocoder 110 will process the signals resulting in a minimal amount of buzzes or pops in the outputted audio to the speaker 114.

To further insure the alternate frequency neighbor searches go completely unnoticed, the DSP 104 may vary the repetition rate of the searches. By way of example, a frst neighboring frequency search may evaluate two neighboring frequencies and take a total search time of about 40 milliseconds. The DSP 104 will then process incoming signals for a period of time until another silence period is detected, and then search a third, fourth, and fifth neighboring frequency. This varying rate reduces the chances that a neighboring frequency search will last longer than a silence period and be audibly detectable by the user. Various repetition rate systems are used currently within Time Division Multiple Access (TDMA) mobile terminals and are well known in the art.

Figure 3 illustrates the steps of the present invention. During a conversation, the mobile terminal 16 receives signals from the base station 12 containing both voice/audio messages and control/data messages. The messages are received by the receiver 102 and sent to the DSP 104 where they are processed (block 300). Upon receiving the search request, the DSP 104 analyzes the incoming and outgoing voice/audio messages (block 310). If the messages do not contain a period of silence (block 320), the next voice/audio frame is analyzed to determine if it contains a period of silence. When a period of silence is detected by the DSP 104, the processing of the voice/audio messages is momentarily stopped as the DSP 104 conducts the neighboring frequency search. During this pause in message processing, background noise stored in memory 107 may be forwarded to the vocoder 110 to prevent audible discrepancies that are output to the speaker (block 330).

To perform the search, the DSP 104 changes from the current frequency on which it is receiving and transmitting signals. The receiver 102 receives signals on the neighboring frequency list and channel quality measurements, such as strength, BER, WER, etc. (block 340). Once the neighboring frequencies have been tested, the receiver is retuned to the original CDMA frequency and incoming and outgoing signal processing is resumed (block 350). The channel quality measurements are forwarded to the base station 12 which makes the determination when to handoff the active call to one of the neighboring base stations (block 360).

The present invention has been described in terms of a CDMA network. However, the present invention is also applicable to other networks including TDMA and Frequency Division Multiple Access (FDMA). These other networks include a signal processing unit that process incoming and outgoing messages. The signal processing units are equipped with a means to determine a period of silence within the incoming and outgoing messages that provides a pause to perform a search of neighboring base stations. The search may be performed and signal processing resumed without a noticeable interrupt in incoming and outgoing voice/audio messages.

Certain modifications and improvements will occur to those skilled in the art upon a reading of the foregoing description. It should be understood that all such modifications and improvements have been deleted herein for the sake of conciseness and readability but are properly within the scope of the following claims.

Claims

What is claimed is:

1. A method of a mobile terminal searching signal strength of neighboring base stations within a CDMA cellular communications network comprising: receiving a signal at a mobile terminal from a first base station, the signal being transmitted at a first frequency; processing the signal at a signal processing unit; determining a period of silence within the signal; thereafter, stopping the processing of the signal and searching at least one alternate frequency of at least one neighboring base station; and after searching the at least one alternate frequency, restoring signal processing of the signal through the first base station at the first frequency.

2. The method of claim 2, wherein an audio signal output to a user is seamless without interruptions.

3. The method of claim 2, further including sending stored audio to a vocoder while searching the alternate frequencies.

4. The method of claim 1 , further including receiving a request for searching the alternate frequencies prior to determining the period of silence.

5. The method of claim 1 , wherein the period of silence includes both incoming and outgoing messages.

6. The method of claim 1 , wherein the neighboring base station searching uses varying repetition rates.

7. The method of claim 1 , wherein periods of silence are caused by natural pauses within human speech.

8. A method of performing alternate frequency neighbor searches within a CDMA wireless communications network comprising the steps of: receiving signals at a mobile terminal at a first frequency; processing the signals at a digital signal processor; sending the processed messages to a vocoder; determining a period of silence within the received signals and stopping the sending of the processed messages to the vocoder; searching at least one alternate base station operating at a second frequency and performing a strength measurement; and after searching the alternate base station, sending processed messages to the vocoder

9. The method of claim 8, wherein sending processed messages to the vocoder after searching the alternate frequencies occurs before the period of silence expires thereby resulting in a seamless audio output.

10. The method of claim 8, further including sending previous audio messages to the vocoder while searching the at least one alternate base station.

11. The method of claim 8, wherein the step of determining a period of silence includes determining silence in both the received signals and outgoing signals.

12. A CDMA wireless communication network method of performing alternate frequency testing comprising the steps of: receiving an incoming signal at a receiver; processing the incoming signal at a digital signal processor; sending the processed incoming signal from the digital signal processor to a vocoder; processing outgoing messages at the digital signal processor; determining a period of silence within the incoming and outgoing signals at the digital signal processor; performing an alternate frequency test; and after completing the alternate frequency test, sending the incoming signals from the digital signal processor to the vocoder and processing the outgoing messages at the digital signal processor.

13. The method of claim 12, wherein the step of determining a period of silence within the incoming and outgoing signals is performed after receiving a request for testing an alternate frequency.

14. The method of claim 12, wherein during the alternate frequency search, stopping the sending of the processed incoming signals from the digital signal processor to the vocoder and the processing of outgoing messages at the digital signal processor.