US20090080423A1

US20090080423A1 - Systems and methods for adaptively adjusting codec rates for communication networks

Info

Publication number: US20090080423A1
Application number: US12/237,158
Authority: US
Inventors: David B. Ewing
Original assignee: Individual
Current assignee: Synapse Wireless Inc
Priority date: 2007-09-24
Filing date: 2008-09-24
Publication date: 2009-03-26
Also published as: US7852820B2; US20090098898A1; US20090080455A1

Abstract

The present disclosure generally relates to systems and methods for adaptively controlling codec rates based on a quality of a communication link such that the codec rate used for data communicated over the communication link automatically changes as the quality of the communication link changes. In one exemplary embodiment, a high codec rate is enabled if the quality of the communication link is high. If a parameter indicative of a quality of the communication link falls below a threshold, the codec rate is decreased. Decreasing of the codec rate generally decreases the quality of the encoded voice data that is to be routed through the network. However, decreasing of the codec rate also decreases the amount of encoded data that is to be routed through the network. Therefore, when the codec rate is decreased, a higher percentage of the encoded data provided by the codec is likely to be received by an end device that is playing the voice data to a user resulting in an improved voice message.

Description

CROSS REFERENCE TO RELATED APPLICATION

This application claims priority to U.S. Provisional Patent Application No. 60/974,836, entitled “Wireless Communication Networks,” and filed on Sep. 24, 2007, which is incorporated herein by reference.

RELATED ART

Wireless communication poses many significant challenges that must be addressed if robust and reliable communication is to be achieved. In this regard, wireless signals are susceptible to noise from various interferers, and the interference encountered by wireless signals within a wireless network can drastically change over time. Indeed, external interferers can generate significant noise that fluctuates. Further, within a wireless network, multiple nodes may attempt communication at the same time causing data collisions. Weather can even affect communication performance. Moreover, the quality and throughput of communication links in wireless networks is constantly changing due to many factors. At times, such quality and throughput may fall below acceptable levels.
When data being routed through a network is lost, the data is often retransmitted. However, the retransmission of data is some applications is not feasible. For example, in voice communication, there is typically very little delay between communication of the voice data through a network and playing of the voice data to a user at an end device, such as a telephone. If voice data is lost during network routing, there is often insufficient time to initiate and complete a retransmission of the lost voice data before the voice data is needed at the end device. Accordingly, if significant portions of a voice message are lost during network routing, there is typically insufficient time to retransmit the lost voice data resulting in a voice message that, when played by the end device, is unintelligible and/or garbled. Techniques for increasing the reliability and quality of voice communication through a wireless network are generally desired.

BRIEF DESCRIPTION OF THE DRAWINGS

The disclosure can be better understood with reference to the following drawings. The elements of the drawings are not necessarily to scale relative to each other, emphasis instead being placed upon clearly illustrating the principles of the disclosure. Furthermore, like reference numerals designate corresponding parts throughout the several views.

FIG. 1 is a block diagram illustrating an exemplary wireless network in accordance with the present disclosure.

FIG. 2 is a block diagram illustrating an exemplary node of a wireless network, such as is depicted by FIG. 1.

FIG. 3 is a block diagram illustrating an exemplary network node, such as is depicted in FIG. 2.

FIG. 4 is a flow chart illustrating an exemplary method for adaptively changing the codec rate for a network node, such as is depicted in FIG. 3.

DETAILED DESCRIPTION

The present disclosure generally relates to systems and methods for adaptively controlling codec rates based on a quality of a communication link such that the codec rate used for data communicated over the communication link automatically changes as the quality of the communication link changes. In one exemplary embodiment, a high codec rate is enabled if the quality of the communication link is high. If a parameter indicative of a quality of the communication link falls below a threshold, the codec rate is decreased. Decreasing of the codec rate generally decreases the quality of the encoded voice data that is to be routed through the network. However, decreasing of the codec rate also decreases the amount of encoded data that is to be routed through the network. Therefore, when the codec rate is decreased, a higher percentage of the encoded data provided by the codec is likely to be received by an end device that is playing the voice data to a user resulting in an improved voice message (e.g., more intelligible or less garbled). If the quality of the communication link improves such that the monitored parameter rises above the threshold, then the codec rate is increased to increase the quality of the encoded voice data. In such a situation, the quality of the communication link has improved such that it can better handle the increased data resulting from the higher codec rate.
FIG. 1 depicts a communication network 20 in accordance with an exemplary embodiment of the present disclosure. As shown by FIG. 1, the network 20 has a plurality of nodes 25. The nodes 25 can be stationary or mobile. In one exemplary embodiment, the nodes 25 communicate with one another via wireless signals, but if desired, any of the nodes 25 may be coupled to any of the other nodes 25 and communicate via a physical medium.
In one exemplary embodiment, the network 20 is configured as a mesh network in which any of the nodes 25 may communicate directly or indirectly with any of the other nodes 25. In addition, the nodes 25 communicate wireless signals, such as radio frequency (RF) signals or signals in other frequency bands, according to I.E.E.E. 802.15.4 or other types of known protocols. Other types of networks may be employed in other embodiments. Various wireless networks are described in U.S. Provisional Patent Application No. 60/953,630, entitled “Sensor Networks,” and filed on Aug. 2, 2007, which is incorporated herein by reference. Various wireless networks are also described in U.S. Provisional Patent Application No. 61/099,453, entitled “Systems and Methods for Controlling Wireless Sensor Networks,” and filed on May 2, 2008, which is incorporated herein by reference. Wireless networks are further described in U.S. patent application Ser. No. 12/114,566, entitled “Systems and Methods for Dynamically Configuring Node Behavior in a Sensor Network,” and filed on May 2, 2008, which is incorporated herein by reference. As will be described in more detail hereafter, voice data and/or other types of data, such as sensor data, can be routed through the nodes 25 of the network 20.
As described in U.S. Provisional Application No. 60/953,630, repeaters (not shown) may be used to extend the communication range of the network 20. In addition, any of the nodes 25 may similarly regenerate signals and, therefore, function as a repeater.
Note that each node 25 is associated with a unique identifier, sometimes referred to as a “node address,” that uniquely identifies such node from other nodes in the network 20. Any signal destined for a node preferably includes the node's unique identifier so that any node receiving the signal can determine whether it is the signal's destination. If it is the destination, then the node responds to the signal as appropriate. For example, if a message identifying a particular node 25 defines a command to perform an action, then the identified node 25, upon receiving the signal, may be configured to further process the signal based on the node identifier and to thereafter perform the commanded action.
In one exemplary embodiment, the network 20 is packet-based. Each data packet has a header, which includes various control information, such as the identifiers of the node or nodes that are to respond and process the packet, and a data portion, which includes payload data, such as voice data or other types of data. The packets may be communicated via any desired protocol. Note that more than one node identifier may be included in the header of a packet. For example, in one embodiment, the node identifier of the ultimate destination of the packet and the source of the packet are included in the header. In addition, if the packet is to hop through intermediate nodes before being received at its ultimate destination, the header includes the node identifier for the next hop (i.e., the next node 25 that is to receive the packet) within the network 20.
When a message is transmitted from a node 25, referred to hereafter as the “transmitting node,” to another node, referred to hereafter as the “receiving node,” the receiving node 25 transmits an acknowledgement back to the transmitting node 25 indicating that the message has been received. If the transmitting node 25 does not receive such an acknowledgement within a certain time period after transmitting the message, then the transmitting node 25 assumes that the message has not been successfully received and attempts to retransmit the message. The transmitting node 25 will continue retransmitting the message until it receives an acknowledgement for the message or once a predefined number of retransmissions have been attempted. The use of acknowledgements to enhance the robustness and reliability of a network is generally well-known and will not be described in detail herein.
FIG. 2 depicts a node 25 in accordance with an exemplary embodiment of the present disclosure. As shown by FIG. 2, the node 25 has control logic 311 for generally controlling the operation of the node 25. The control logic 311 can be implemented in software, hardware, firmware, or a combination thereof. In an exemplary embodiment illustrated in FIG. 2, the control logic 311 is implemented in software and stored in memory 314.
Note that the control logic 311, when implemented in software, can be stored and transported on any computer-readable medium for use by or in connection with an instruction execution apparatus that can fetch and execute instructions. In the context of this document, a “computer-readable medium” can be any means that can contain, store, communicate, propagate, or transport a program for use by or in connection with the instruction execution apparatus.
The exemplary embodiment of the node 25 depicted by FIG. 2 comprises at least one conventional processing element 323, such as a digital signal processor (DSP) or a central processing unit (CPU), that communicates to and drives the other elements within node 25 via a local interface 326, which can include at least one bus. Furthermore, a data interface 329, such as a USB port or RS-232 port, allows data to be exchanged with external devices.
The node 25 also has a network interface 334 for enabling the control logic 311 to communicate with other nodes 25. In one exemplary embodiment, the interface 334 is configured to communicate wireless signals, but the interface 334 may communicate with another node 25 via a physical medium, if desired.
As shown by FIG. 2, the network interface 334 has an antenna 336, a transceiver 337, and a protocol stack 339. The stack 339 controls the communication of data between the network interface 334 and the other nodes 25. In one exemplary embodiment, the stack 339 is implemented in software. However, in other embodiments it is possible for the stack 339 to be implemented in hardware, software, firmware, or a combination thereof.
As shown by FIG. 2, the node 25 comprises various user interface devices for enabling information to be exchanged with a user. For example, the node 25 comprises a user input interface 344, such as a keypad, buttons, and/or other types input devices, for enabling a user to enter data or otherwise provide inputs. The node 25 also has a user output interface 347, such as a liquid crystal display device (LCD), for displaying or otherwise indicating information to a user. In addition, the node 25 has a microphone 352 for sensing audible sounds, such as voice, and a speaker 355 for generating audible sounds. Other types of user interface devices may be employed in other embodiments. The node 25 also comprises a clock 363 for enabling the control logic 311 to track time, as may be desired.
As described above, voice data is preferably encoded before being packetized and transmitted. Each node 25 preferably comprises at least one codec for encoding voice data before packetization of the voice data. The encoding algorithm compresses the voice data such that the encoded data is smaller than the voice data prior to coding. Further, a codec in a receiving node decodes and thereby decompresses the voice data before the voice data is played via a speaker. Known or future-developed techniques for performing encoding/decoding of voice data may be employed.
In one exemplary embodiment, a node 25 comprises a coding element 500, which can encode voice data at any of various codec rates. In the exemplary embodiment shown by FIG. 2, the coding element 500 has a plurality of codecs 501, 502. Each codec 501, 502 is configured to perform a different coding algorithm relative to the other codec. For example, one codec 501, 502 may encode voice data at one rate, and the other codec 501, 502 may encode voice data at another rate. The codec that encodes at a higher rate produces more data when encoding a given set of voice data relative to the codec that encodes at the lower rate. For the purpose of illustration, assume that codec 501 encodes voice data at 9600 Baud and that codec 502 encodes data at a lower rate, such as 4800 Baud. In other examples, other rates are possible.
As shown by FIG. 3, the node's microphone 352 is coupled to an analog-to-digital (A/D) converter 509. The microphone 352 generates analog audio signals based on sounds, such as voice, detected by the microphone 352, and the A/D converter 509 converts the audio signals into digital data, referred to as “voice data.” The A/D converter 509 is coupled to both codecs 501, 502 and transmits the voice data to both codecs 501, 502. The control logic 311 selectively enables one of the codecs 501, 502 and disables the other. Thus, only one of the codecs 501, 502 at any given time encodes the voice data to provide encoded data that is provided to the network interface 334 for transmission to a remote node 25 or nodes 25. Accordingly, at any given time, the network interface 334 receives encoded voice data from only one of the codecs 501, 502. Note that each of the codecs 501, 502, when encoding voice data, preferably performs a respective compression algorithm.
Upon receiving encoded voice data, the network interface 334 packetizes the encoded voice data and wirelessly transmits data packets containing the encoded voice data to at least one remote node 25. Note that the network interface 334 may also receive other types of data from a data source, such as the data interface 329 and/or the user input device 344, and transmit such data along with the encoded voice data.
For example, in one exemplary embodiment, the stack 339 is configured to packetize voice data from the codecs 501, 502 into a plurality of data packets, referred to hereafter as “voice packets,” and to packetize data from the data interface 329 or other data source into a plurality of data packets, referred to hereafter as “non-voice packets.” The stack 339 then interleaves the voice packets and non-voice packets for transmission by the transceiver 337. The extent and manner of interleaving may depend on various factors, such as the amount of voice data that is ready for communication and the amount of data from the data source that is ready for communication. In one exemplary embodiment, the stack 339 prioritizes the voice data, which often needs to be transmitted in real-time for many applications, such that the throughput of the voice data remains above a specified threshold when there is voice data present for communication. Thus, to keep the throughput of the voice data above the threshold, the transmission of data from the data source may be delayed.
The control logic 311 is configured to adaptively select one of the codecs 501, 502 for encoding data based on various factors. In one exemplary embodiment, the control logic 311 selects between the codecs 501, 502 based on a parameter indicative of the current quality of the wireless communication channel used by the transceiver 337 for the voice packets. In particular, when communicating with one or more remote nodes 25, the stack 339 monitors the communication to determine a communication quality parameter indicative of the current quality of the network link between the transceiver 337 and the remote node 25 or nodes 25.
Note that there are various parameters that may be calculated or otherwise determined to indicate link quality. In one exemplary embodiment, the stack 339 calculates the throughput of the communication link based on the acknowledgments received from other nodes 25. In this regard, as described above, the network interface 334 is configured to receive acknowledgments or other messages indicating which of the packets transmitted by the transceiver 337 over the network 20 have been successfully received by the remote node 25. The stack 339, based on such acknowledgments, calculates a value, referred to hereafter as the “throughput value,” indicative of the link's throughput. For example, the throughput value may be a ratio of the number of data packets successfully transmitted to the total number of data packets transmitted during a particular time period. In this regard, the stack 339 tracks the total number of packets transmitted by the transceiver 337 from the interface 334 to the remote node 25 over some time period, such as one minute, for example. Based on the acknowledgements received from the other nodes 25, the stack 339 determines the number (referred to hereafter as the “received number”) of such packets successfully received by the other nodes 25. The stack 339 divides the received number by the total number to calculate the throughput value, and the stack 339 transmits the throughput value to the control logic 311. Thus, the throughput value is a ratio of the number of successfully transmitted packets to the total number of transmitted packets such that a higher throughput value generally indicates better link quality. Other techniques for calculating the throughput value or other values indicative of link quality are possible in other embodiments.
The control logic 311 is configured to select which of the codecs 501, 502 that is to be currently enabled based on the communication quality parameter. In the instant example, assume that the communication quality parameter received from the network interface 334 is the throughput value described above for which a higher value indicates that a higher percentage of data packets are being successfully received and that the quality of communication is, therefore, better. The control logic 311 is configured to compare the throughput value to a predefined threshold. When the throughput value is high so that it exceeds the predefined threshold, the control logic 311 is configured to enable the codec 501 that encodes at a higher rate in an effort to improve the quality of the voice communication. In such a case, the control logic 311 disables the other codec 502. However, if the throughput value falls below the threshold indicating that the network link used by the transceiver 337 has degraded, the control logic 311 enables the codec 502 that encodes at a lower rate and disables the other codec 501. By selecting the lower-rate codec 502, the encoded voice data provided to the network interface 334 is of a lower quality than would have otherwise been provided by the higher rate codec 501, but there is less encoded data that must be transmitted to the remote node 25. Thus, it is less likely that the degraded communication occurring between the network interface 334 and the remote node 25 will cause a given portion of the voice communication to become unintelligible. Accordingly, the control logic 311 is able to react to changing communication conditions on the network channel used by the transceiver 337 in order to select a more suitable encoding algorithm for the voice data.
In the embodiment shown by FIG. 3, only two codecs 501, 502 are shown. However, in other embodiments, other numbers of codecs may be employed and similar techniques may be used to selectively enable the suitable codec for encoding the voice data from the microphone 352. In addition, it is possible for any of the codecs to be implemented in hardware, software, or a combination therefore. In one exemplary embodiment, the codecs 501, 502 are implemented in software and stored on a computer-readable medium, but other configurations of the codecs 501, 502 are possible in other embodiments.
In one exemplary embodiment, the stack 339 utilizes features commonly used in various communication protocols, such as real-time protocol (RTP), to provide monitoring and real-time control of the packets communicated by the network 20. For example, the nodes 25 are preferably synchronized, and the stack 339 of each node 25 inserts a time stamp in the header of each packet transmitted by the node 25. The time stamp indicates the packet's time of transmission. The time stamps can be used by a receiving node to reorder packets that are received out of order and to monitor various communication parameters, such as jitter and latency.
An exemplary use and operation of a node 25 in adaptively controlling its codec rate will be described in more detail hereafter.
Assume that the user of one node 25, referred to as the “transmitting node,” is to convey voice messages to another node 25, referred to as the “receiving node,” that is in direct (no repeaters) communication with the transmitting node 25. For illustrative purposes, assume that the codec 501 of the transmitting node 25 encodes/decodes data at one rate, such as 9600 Baud, and that codec 502 of the transmitting node 25 encode/decodes data at a lower rate, such as 4800 Baud. Other rates are possible in other examples.
As shown by block 411 of FIG. 4, the stack 339 of the transmitting node 24 calculates a link quality parameter value (LQP) value indicative of the quality of the channel used for communication between the transceiver 337 and the receiving node 25. As an example, the LQP value may be the throughput value described above calculated for the communication occurring during a particular time period, such as one minute for example. In the current example, a higher LQP value indicates a higher quality for the communication channel.
As shown by block 415 of FIG. 4, the control logic 311 of the transmitting node 25 compares the LQP value to a threshold (“TH”). In one exemplary embodiment, the threshold is established based on the following criteria. In this regard, if the threshold is exceeded, then the quality of the communication channel is high enough such that it is likely that encoded data provided by the codec 501 can be transmitted over the channel with acceptable voice quality if the voice data is played at the remote node. If the threshold is not exceeded, then the communication channel has become sufficiently degraded such that it is unlikely that encoded data provided by the codec 501 can be transmitted over the channel with acceptable voice quality due to packets lost by the network 20. As an example, the threshold may be established such that if the threshold is not exceeded, then it is likely that a voice message defined by the encoded data from the codec 501 would be unintelligible or garbled if played at the remote node 25 due to an excessive amount of lost packets in the channel.
For illustrative purposes, assume that the LQP value exceeds the threshold when the user of the transmitting node 25 initiates a communication session or otherwise begins conveying voice messages to the user of the receiving node 25. In such an example, a voice message sensed by the microphone 352 of the transmitting node 25 is converted to digital data by the A/D converter 509. In the instant example, the LQP value exceeds the threshold, and the control logic 311 of the transmitting node 25, therefore, enables codec 501 and disables the codec 502, as shown by block 421 of FIG. 4. Thus, the digital data from the A/D converter 509 is encoded by the codec 501 into a compressed form. Such encoded data is transmitted to the network interface 334, and the stack 339 packetizes the data for transmission to the receiving node 25. The transceiver 337 then wirelessly transmits such packets to the receiving node 25.
After a few minutes of operation, assume that the communication channel becomes significantly degraded such that the current LQP value falls below the threshold. When this occurs, the control logic 311 of the transmitting node 25 switches the coding element 500 to a lower codec rate. In this regard, the control logic 311 makes a “no” determination in block 415. Thus, the control logic 311 enables the codec 502 and disables the codec 501, as shown by block 425 of FIG. 4, effectively switching from the high codec rate of codec 501 to the low codec rate of codec 502. In this regard, the digital data from the A/D converter 509 is encoded by the codec 502 into a compressed form at a rate lower than that for the codec 501. Such encoded data is transmitted to the network interface 334, and the stack 339 packetizes the data for transmission to the receiving node 25. The transceiver 337 then wirelessly transmits such packets to the receiving node 25. Thus, when the communication channel becomes significantly degraded, the codec rate is adaptively switched from the high codec rate of codec 501 to the low codec rate of codec 502.
If the communication later improves such that the LQP value rises above the threshold, then the control logic 311 enables codec 501 and disable codec 502 adaptively switching to the higher codec rate. Accordingly, the codec rate employed by the transmitting node 25 for the channel between the transmitting node 25 and the receiving node 25 is adaptively updated based on changing conditions that affect the link quality.
In exemplary embodiments described above, multiple codecs are used to enable selection among multiple codec rates. Note that it is possible for a single codec to be employed if such a codec allows its codec rate to be selectively and automatically changed by the control logic 311. Other changes and modifications would be apparent to one of ordinary skill in the art upon reading this disclosure.

Claims

1. A network node, comprising:

a transceiver;

a coding element having at least one codec, the coding element configured to receive voice data and to encode said voice data thereby providing encoded data at a first rate;

a protocol stack configured to packetize the encoded data thereby providing a plurality of data packets to be communicated through a channel of a network;

a transceiver configured to transmit the data packets through the channel; and

logic configured to determine a value indicative of a current quality of the channel, the logic configured to adaptively update the coding element based on the value such that the coding element provides the encoded data at a second rate that is different than the first rate.

2. The network node of claim 1, wherein the coding element comprises a first codec and a second codec, the first codec configured to encode data at the first rate and the second codec configured to decode data at the second rate, and wherein the logic is configured to enable the second codec and to disable the first codec based on the value.

3. The network node of claim 1, wherein the logic is configured to compare the value to a threshold.

4. The network node of claim 1, wherein the coding element comprises a first codec and a second codec, the first codec configured to encode data at the first rate and the second codec configured to decode data at the second rate, wherein the logic is configured to adaptively select, based on the value, one of the first and second codecs for encoding the voice data.

5. The network node of claim 1, wherein the logic is configured to determine the value based on acknowledgements received by the network node from other nodes of the network.

6. A network node, comprising:

a first codec configured to encode data at a first rate;

a second codec configured to encode data at a second rate that is different than said first rate;

a protocol stack;

a transceiver configured to communicate data packets through a channel of a network; and

logic configured to determine a value indicative of a current quality of the channel, the logic configured to adaptively select one of the first and second codecs for encoding voice data based on the value, wherein the selected codec is configured to encode the voice data thereby providing encoded data and to provide the encoded data to the stack, and wherein the protocol stack is configured to packetize the encoded data into a plurality of data packets for transmission by the transceiver through the channel.

7. The network node of claim 6, wherein the logic is configured to compare the value to a threshold.

8. The network node of claim 6, wherein the logic is configured to determine the value based on acknowledgements received by the network node from other nodes of the network.

9. A method for adaptively changing codec rates based on varying conditions for a network channel, comprising the steps of:

receiving voice data;

encoding the voice data for transmission through a network thereby providing encoded data;

packetizing the encoded data thereby providing a plurality of data packets;

transmitting the data packets through a channel of the network;

determining a value based on a current quality of the channel; and

adaptively adjusting a rate of the encoding step based on the value.

10. The method of claim 9, further comprising the step of comparing the value to a threshold, wherein the adjusting step is based on the comparing step.

11. The method of claim 9, further comprising the step of receiving acknowledgements from nodes of the network, wherein the determining step is based on the received acknowledgements.

12. The method of claim 9, wherein the encoding step is performed by a first codec and a second codec, and wherein the method further comprises the step of adaptively selecting one of the first and second codecs for encoding the voice data based on the value, wherein the adjusting step is based on the selecting step.

13. A method for adaptively changing codec rates based on varying conditions for a network channel, comprising the steps of:

receiving voice data;

encoding, via a first codec, a first portion of the voice data for transmission through a network thereby providing first encoded data;

packetizing the first encoded data thereby providing a first plurality of data packets;

transmitting the first plurality of data packets through a channel of the network;

determining a value indicative of a quality of the channel during the transmitting step;

comparing the value to a threshold;

adaptively selecting, based on the comparing step, a second codec for encoding a second portion of the voice data for transmission through the network;

encoding, via the second codec, the second portion of the voice data thereby providing second encoded data;

packetizing the second encoded data thereby providing a second plurality of data packets; and

transmitting the second plurality of data packets through the channel.

14. The method of claim 13, further comprising the step of receiving acknowledgements from nodes of the network, wherein the determining step is based on the received acknowledgements.