CA2300495C - Technique for multi-rate coding of a signal containing information - Google Patents
Technique for multi-rate coding of a signal containing information Download PDFInfo
- Publication number
- CA2300495C CA2300495C CA002300495A CA2300495A CA2300495C CA 2300495 C CA2300495 C CA 2300495C CA 002300495 A CA002300495 A CA 002300495A CA 2300495 A CA2300495 A CA 2300495A CA 2300495 C CA2300495 C CA 2300495C
- Authority
- CA
- Canada
- Prior art keywords
- representation
- packets
- packet
- representations
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/16—Vocoder architecture
- G10L19/18—Vocoders using multiple modes
- G10L19/24—Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
Abstract
In a communications system, multi-rate coding in accordance with the invention is implemented to generate multiple representations of an audio signal at different rates. These representations contain equivalent and/or various amounts of audio information. In an illustrative embodiment, at least one of the representations is a core representation containing core audio information. The remaining representations are enhancement representations containing enhancement audio information. The core representation is necessary for recovering the audio signal with minimal acceptable quality. Such quality is enhanced when the core representation, together with one or more of the enhancement representations, is used to recover the audio signal.
Description
TECHNIQUE FOR MULTI-RATE CODING OF A SIGNALCONTAINING
INFORMATION
Field of the Invention The invention relates to systems and methods for communications of a signal containing information, and more particularly to communications systems and methods for coding the signal to generate multiple representations thereof.
Background of the Invention Communications of audio information play an important role in multimedia applications, and Internet applications such as a music-on-demand service, music preview for online compact disk (CDy purchases, etc. To efficiently utilize bandwidth to communicate audio information, a perceptual audio coding (PAC) technique has been developed. For details on the PAC technique, one may refer to U.S. Patent No. 5,285,498 issued February 8, 1994 to Johnston; and U.S.
Patent No. 5,040,217 issued August 13, 1991 to Brandenburg et al.
In accordance with such a PAC technique, each of a succession of time domain blocks of an audio signal representing audio information is coded in the frequency domain. Specifically, the frequency domain representation of each block is divided into coder bands, each of which is individually coded, based on psycho-acoustic criteria, in such a way that the audio information is significantly compressed, thereby requiring a smaller number of bits to represent the audio information than would be the case if the audio information were represented in a more simplistic digital format, such as the PCM format.
For example, in providing the aforementioned music-on-demand service, a server connected to the Internet may store PAC compressed versions of each available musical P. Kroon 7-9 piece to serve client needs. Each version of the musical piece corresponds to a different connection speed at which a client, e.g., a personal computer (PC) having a modem, can afford to communicate over the Internet. The quality, or the lack of distortion, of the version of the musical piece increases with the connection speed corresponding thereto. Thus, for instance, if the server supports (a) a plain old telephone service (POTS) connection speed of about 28.8 kb/sec, (b) an integrated services digital l~ network (ISDN) connection speed of about 64 kb/sec, and (c) a dual ISDN connection speed on the order of 100 kb/sec, three corresponding versions of the musical piece having the respective qualities need to be stored in the server.
However, the storage of musical pieces in this manner is undesirably inefficient and occupies much memory space especially when a large number of musical pieces need to be made available.
In delivering the service to a client at a given connection speed, the server may packetize the 2~ corresponding audio information in the storage, and communicate the resulting packets through a packet switched network, e.g., the Internet. However, in the event that some of the packets are lost in transit because of imperfect network or channel conditions, which is likely, 2~ the quality of the received signal representative of a musical piece would be significantly degraded.
Accordingly, there exists needs for efficiently storing and distributing information at different rates, and effectively maintaining the minimum acceptable quality 3~~ of the received signal despite imperfect network or channel conditions.
Summary of the Invention In accordance with the invention, mufti-rate coding is implemented to generate multiple subrate representations of P. Kroon 7-9 3 a signal containing information, e.g., audio information.
These representations are different from one another and may be delivered at rates lower than or equal to the required delivery rates of the information.
For example, in providing the music-on-demand service described above, at least one of the subrate representations, referred to as a "C-representation," may contain core information delivered at a subrate of 28.8 kb/sec. The other subrate representations, referred to as 1 ~ a "E1-representat:ion" and "E~-representation, each may contain enhancement information delivered at a subrate of 36 kb/sec. Because of the design of the mufti-rate coding in accordance with the invention, in this instance recovery of the signal based on the C-representation alone affords l.~ the minimum acceptable 28.8 kb/sec signal quality; recovery of the signal based on the C-representation in combination with either E~-representation or E~-representation affords a higher 64 kb/s~~c signal quality; and recovery of the signal based on the C-representation in combination with 20 both E;-representation and E~-representation affords the highest 100 kb/s~=c signal quality. Advantageously, the server of the af~~rementioned music-on-demand service needs to store in its memory the subrate representations, i.e., 28.8 kb/sec C-representation, 36 kb/sec E1-representation 25 and 36 kb/sec E~-representation, of each musical piece, in lieu of the 28.8 kb/sec, 64 kb/sec and 100 kb/sec versions thereof as in prior art, to accommodate the different connection speeds and quality requirements, thereby effectively saving the memory space.
30 In accordance with an aspect of the invention, when the subrate representations are communicated to a client terminal in the Form of packets, each packet includes at least an informa~~ion content derived from one of the representations, and an indicator identifying the 3_'~ representation f..om which the information content is derived. Despite losses of some packets in transit because of imperfect channel or network conditions, the signal is recovered based on at least the received packets indicated to contain C-representation information to maintain the minimum acceptable signal quality.
In accordance with one aspect of the present invention there is provided an apparatus for processing a signal comprising: a coder for generating at least first and second representations of the signal, the first and second representations being different from each other; and a controller for packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
In accordance with another aspect of the present invention there is provided a method for processing a 2S signal comprising: generating at least first and second representations of the signal, the first and second representations being different from each other; and packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the 4a information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
Brief Description of the Drawings Fig. 1 illustrates an arrangement embodying the principles of the invention for communicating audio information through a communication network;
Fig. 2 is a block diagram of a server in the arrangement of Fig. 1;
Fig. 3A illustrates a homogeneous multidimensional lattice based on which a prior art quantizer performs quantization;
Fig. 3B illustrates a first non-homogeneous multidimensional lattice based on which a first complementary quantizer in the server of Fig. 2 performs quantization;
Fig. 3C illustrates a second non-homogeneous multidimensional lattice based on which a second complementary quantizer in the server of Fig. 2 performs quantization;
Fig. 4 illustrates a stream of packets generated by the server of Fig. 2; and Fig. 5 is a flow chart depicting the steps whereby a client terminal in the arrangement of Fig. 1 processes the packets from the server.
P. Kroon 7-9 5 Detailed Description Fig. 1 illustrates arrangement 100 embodying the principles of the invention for communicating information, e.g., audio information. In this illustrative embodiment, server 105 in arrangement 100 provides a music-on-demand service to client terminals through Internet 120. One such client terminal is numerically denoted 130 which may be a personal computer (PC). As is well known, Internet 120 is a packet switched network for transporting information in packets in accordance with the standard transmission control protocol/Internet protocol (TCP/IP).
Conventional software including browser software, e.g., the NETSCAPE NAVIGATOR or MICROSOFT EXPLORER browser is installed in client terminal 130 for communicating information with server 105, which is identified by a predetermined uniform resource locator (URL) on Internet 120. For example, to request the music-on-demand service provided by server 105, a modem lnot shown) in client terminal 130 is used to first establish communication connection 125 with Internet 120. Depending on the telecommunication facility subscribed by the user of client terminal 130, communication connection 125 may be limited by different connection speeds. For instance, a plain old telephone service (POTS) connection typically affords a 2~ connection speed of about 28.8 kb/sec; an integrated services digital network (ISDN) connection typically affords a connection speed of about 64 kb/sec; and a dual ISDN connection typically affords a connection speed on the order of 100 kb/sec.
After the establishment of communication connection 125, in a conventional manner, client terminal 130 is assigned an IP address for its identification. The user at client 130 may then access the music-on-demand service at the predetermined URL identifying server 105, and request a P. Kroon 7-9 selected musical piece from the service. Such a request includes the IP ;address identifying client terminal 130, and its connecti«n speed.
In prior art, in providing the music-on-demand service, a server needs to store versions of each musical piece corresponding to different connection speeds supported by the server. The audio quality (distortion) of a version of the musical piece increases (decreases) with the corresponding connection speed. Thus, if a prior art server supports three connection speeds, e.g., 28.8 kb/sec, 64 kb/sec and 100 kb/sec, the server needs to store three different versions of each musical piece available having the respective qualities. However, the storage of musical pieces in this manner is undesirably inefficient and occupies much memory space especially when a large number of musical pieces need to be made available. In addition, in delivering the service to a client terminal, the server typically sends the audio information in the form of packets through the Internet. However, in the event that 2~ some of the packets are lost in transit because of imperfect network or channel conditions, which is likely, the quality of the received audio information would be significantly degraded..
In accordance with the invention, mufti-rate audio coding is implemented in server 105 to generate subrate representations of each musical piece to save memory space.
Different combinatior.,s of the subrate representations of a musical piece correspond to different connection speeds, and audio qualities of the musical piece. In general, the more subrate representations are communicated to a client terminal, the higher t:he audio quality of the musical piece recovered at the terminal and, of course, the higher the connection speed required of the terminal. For example, in this illustrative embodiment, three subrate representations are used in server 10_'> to serve each musical piece in P. Kroon 7-9 accordance with the invention. One of the subrate representations represents core audio information contained in the musical piece, and is referred to as a "C-representation." The other two subrate representations represent first and second enhancement audio information contained in the musical piece, and are referred to as "E:-representation" and "E~-representation," respectively.
Because of the design of the mufti-rate coding in accordance with the invention, the audio signals recovered based on the C-representation alone, although viable, afford the minimum acceptable quality version of a musical piece; the audio signals recovered based on the C-representation in combination with either E1-representation or E_-representa~ion afford a relatively high quality version of the musical. piece; the audio signals recovered based on the C-representation in combination with both E,-representation and E_.-representation afford the highest quality version of they musical piece. However, any audio signals recovered based only on the E1-representation and/or E---representations are not viable.
An embedded audio coder in accordance with the invention is used in server 105 to generate the C-representation requiring a bit rate of, say, 28.8 kb/sec for communication thereof; the E,.-representation requiring a bit rate of, say, 3Ei kb/sec; and the E~-representation requiring a bit rate c>f, say, 36 kb/sec as well. These bit rates are selected such that if all of the representations are used, the quality of the recovered musical piece version is close to that of a 100 kb/sec version generated by a conventional non--embbeded audio coder. Similarly, the quality of the recovered musical piece version based on a combination of the C-representation with the E1-representation or E~-representatio_n is close to that of a 64 kb/sec version generated by the conventional non-embedded audio c:oder. Apparently, the quality of the recovered musical piece version based on the C
P. Kroon 7-9 representation alone is the same as that of a 28.8 kb/sec version generated by the conventional non-embedded audio coder. Advantageously, server 105 only needs to store in its memory the 28.8 kb/sec C-representation, 36 kb/sec E:-representation and 36 kb/sec E~-representation of each musical piece, in lieu of the 28.8 kb/sec, 64 kb/sec and 100 kb/sec versions thereof as in prior art, to accommodate different connection speeds (e. g., 28.8 kb/sec, 64 kb/sec and 100 kb/sec), thereby saving the memory space.
The aforementioned embbeded audio coder implementing multi-rate coding in accordance with the invention will now be described. Fig. 2 illustrates one such embbeded audio coder, denoted 203, in server 105. An analog signal a(t) representing a musical piece is fed to embedded audio coder 203 in providing the music-on-demand service. In response to such an analog signal, analog-to-digital (A/D) convertor 205 in coder 203 digitizes a(t) in a conventional manner, providing PCM samples of a(t). These PCM samples are fed to both filterbank 209 and perceptual model processor 211.
2~ Filterbank 209 divides the samples into time domain blocks, and performs a modified discrete cosine transform (MDCT) on each block to provide a frequency domain representation therefor. Such a frequency domain representation is bandlimited by low-pass filter (LPF) 213 2~ to the 0 to 10 kHz frequency range in this instance. The resulting MDCT coefficients are grouped by quantizer 215 according to coder bands for quantization. These coder bands approximate the well known critical bands of the human auditory system, although limited to the 0 to 10 kHz 30 frequency range in this instance. Quantizer 215 quantizes the MDCT coefficients corresponding to a given coder band with the same quantizer stepsize.
Perceptual model processor 211 analyzes the audio signal samples and determines the appropriate level of 3~ quantization (i.e., stepsize) for each coder band. This P. Kroon 7-9 level of quantization is determined based on an assessment of how well the audio signal in a given coder band masks noise. Quantizer 215 generates quantized MDCT coefficients for application to loss-less compressor 219, which in this instance performs a conventional Huffman compression process on the quantiz.ed coefficients, resulting in the aforementioned C-representation on lead 261. The output of compressor 219 is fed back to quantizer 215 through rate-loop processor 225. In a conventional manner, the latter adjusts the output of quantizer 215 to ensure that the bit rate of the C-representation is maintained at its target rate, which in this instance is 28.8 kb/sec.
In this illustrative embodiment, the E1-representation and E~-representation are generated by coder 203 for enhancing the quality of the musical piece which contain spectral information concerning relatively high frequency components of the audio signal, e.g., in the 7 to 20 kHz range. To that end, t:he quantized MDCT coefficients from quantizer 215 are subtracted by subtracter 229 from the MDCT output of filterbank 209. The resulting difference signals are duplicated by duplicator 231, and then bandlimited respectively by band-pass filters (BPFs) 223 and 233 to the 7 to 20 kHz range. Each of quantizers 243 and 253 receives a copy of the filtered difference signals and quantizes th.e received signals according to predetermined stepsizes.
Quantizers 243 and 253 may be scalar quantizers or multidimensional quant:izers, and may comprise a complementary qL:antizer pair. Complementary scalar quantizers are well known in the art, and described, e.g., in V. Vaishampayan, "Design of Multiple Description of Scalar Quantizers," IhEE Transactions on Information Theory, Vol. 39, No. 3, May 1993, pp. 821-834. In general, a pair of complementary scalar quantizers may be defined by the following encoding functions f: and f , respectively:
P. Kroon 7-9 10 .f , ~x~ : ~i ~ - x, _,"'_,1 , and where ~ represents the real axis, ml = 2'1 and m2 = 2~'~, where S1 and S2 repre~~ent the bit rates for quantizers 243 and 253, respectively. As is well known, associated with each of the quantized values xi and y~ for f; and f~, respectively, is a range or partition [x, y) on the real axis such that all the values in this range are quantized to x: or y, .
In prior art, to take advantage of the correlation between x-~ and y; from fl and f~ having a complementary relationship, joint decoding, also known as "center decoding, " on (xi, y-) is performed in a de-quantizer to realize the optimum decoded value z~., such that the resulting distortion or quantization error is minimized.
The center decocting function, , performed in the de-quantizer may be expressed as follows:
Ll x y ' x ~~' ml,~=m2 ~ ~Z ~ m ~,y~ k ,=t.~=I k=1 It should be noted that not all (x;, y~) are valid decodable combinations depending upon the overlap between their associated partitions. Let Q,., Q~ and Q- be the c0 average distortions associated with f:, f~ and center decoding function , respectively, and let's assume that f:
and f_ are equivalent, i . a . , S1 = S2 = S . I f Q~_ < 2--~ and Q_ < 2--', by minimizing Q- subject to the condition Q1 and P. Kroon 7-9 11 Q ~ Q, where Q is a predetermined distortion value, it can be shown that tr.e value of Q- is always greater than the following limit:
Q ~ 1 2_,~
That is, use of the complementary scalar quantizers affords at most a 3 dB gain, compared with the case where only an individual scal~ir quantizer is used.
However, it. has been recognized that the average distortion O associated with center decoding can be improved if the complE=_mentary quantizers used are l.0 multidimensiona__, rat)zer than scalar as in prior art. In this illustrative embodiment, quantizers 243 and 253 are complementary multidimensional quantizers in accordance with the invention. Preferably, they are non-homogeneous multidimensional lattice quantizers.
._5 In order to more appreciate the advantages of use of complementary non-homogeneous multidimensional lattice quantizers in accordance with the invention, let's first consider a prior art homogeneous 2-dimensional lattice quantizer using a square lattice in a 2-dimensional region :?0 for quantizatio:n. Fig. 3A illustrates one such 2-dimensional region which is defined by X1 and X2 axes and denoted 360. Region 360 in this instance has a square lattice and contains Voronoi regions or cells, e.g., cells 367 and 369, whose length is denoted D, where D represents ?5 a predetermined value. As shown in Fig. 3A, these cells are homogeneously distributed throughout region 360, and are each identified by a different code. As is well known, in the quantization process, the prior art quantizer assigns to an input ;ample point (xl, x2) the code 30 identifying the cell in which the sample point falls, where P. Kroon 7-9 12 xl E X1 and x2 E X2. For example, sample points having 0 xl < D, and 0 < x2 < D are each assigned the code identifying cell 367. In addition, sample points having D
xl < 2D, and D ~ x2 < 2D are each assigned the code identifying cell 369. In practice, each code assignment is achieved by looking up a codebook.
The above prior art quantizer imposes an average distortion proportional to D= which in turn is proportional to 2--s, where in the multidimensional case here S
represents the number of bits/sample/dimension multiplied by the sample rate.
As mentioned before, in the preferred embodiment, quantizers 243 and 253 are complementary non-homogeneous multidimensional lattice quantizers. For example, in the 2-dimensional case, quantizers 243 and 253 use non-homogeneous rectangular lattices in 2-dimensional regions 370 and 390, respectively. In Fig. 3B, like region 360, region 370 is defined by X1 and X2 axes. However, unlike region 360, region 370 contains Voronoi regions or cells, e.g., cells 367 and 369, which are in different shapes and thus non-homogeneous throughout region 370. By way of example, the vertical boundaries of the rectangular cells in region 370 intersect the Xl axis at xl = 0, 0.5~, 2.0~, 2.5~, 4.0~ ..., with the separations between successive vertical boundaries alternating between 0.5~ and 1.5~. On the other hand, the horizontal boundaries of the rectangular cells in region 370 intersect the X2 axis at x2 - 0, 1.5~, 2.0~, 3.5~, 4.00 ..., with the separations between successive horizontal boundaries alternating between 1.5~ and 0.50. In the quantization process, quantizer 343 assigns to an input sample point (xl, x2) the code identifying the cell in which the sample point falls.
For example, sample paints having 0 < xl < 0.5~, and 0 x2 < 1.5~ are each as~.igned the code identifying cell 377.
P. Kroon 7-9 13 In addition, sample points having 0.5~ < xl < 2.0~, and 1.5~ < x2 < 2.04 are each assigned the code identifying cell 379.
A simple way of designing the rectangular lattice in region 390 of quantizer 253, which is complementary to quantizer 243, is to adopt the vertical and horizontal boundaries in region 370 as the horizontal and vertical boundaries in region 390, respectively. Fig. 3C
illustrates the resulting region 390 containing cells, e.g., cells 391 and 399, which are in different shapes, and thus non-homogeneous throughout region 390. In the quantization process, quantizer 253 assigns to an input sample point (xl, x2) the code identifying the cell in which the sample point falls. For example, sample points having 0 < xl < 1.5~, and 0 ~ x2 < 0.5~ are each assigned the code identifying cell 397. In addition, sample points having 1.5~ < xl < 2.00, and 0.5~ < x2 < 2.0~ are each assigned the code identifying cell 399.
It can be shown that the average distortion for an 2J individual one of quantizers 243 and 253 equals 1.25 a 2--', where a represents a constant which depends on the probability density function of the input signal to the quantizer, and S in this instance equals 36 kb/s. However, stemming from the fact that quantizers 243 and 253 are 2~ complementary quantizers, center decoding on the quantized values from quantizers 243 and 253 respectively can be performed in a de-quantizer. It can be shown that the resulting average distortion Q- associated with 2-dimensional center decoding is no more than 0.25 a 2-='.
3~~ That is, complementary quantizers 243 and 253 when implemented with the 2-dimensional center decoding command a 6 dB improvement in terms of distortion over their scalar counterparts.
The equivalent lattices of three and higher dimensions P. Kroon 7-9 14 of complementary quantizers may be obtained similarly to those of two dimensions described above. However, in three or higher dimensions, it is more advantageous to use a non-homogeneous, non-rectangular (or non-hypercube) lattice in each complementary quantizer.
Referring back to Fig. 2, the quantized signals from quantizer 243 are fed to loss-less compressor 245 which, like compressor 219, achieves bit compression on the quantized signals, resulting in the E;-representation on lead 263. The E,-representation is fed back to quantizer 243 through rate-loop processor 247 to ensure that the bit rate of the E1-representation is maintained at its target rate, which in this instance is S1 = 36 kb/sec.
Similarly, the qu.antized signals from quantizer 253 are fed to loss-less compressor 255 which achieves bit compression on the qua.ntized signals, resulting in the E=-representation on lead. 265. The E~-representation is fed back to quantizer 253 through rate-loop processor 257 to ensure that the bit rate of E~-representation is maintained at its target rate, which in this instance is S2 = 36 kb/sec.
Leads 261, 263 and 265 extend to storage 270 where the C-representation on lead 261 is stored in memory space 271, the E1-representation on lead 263 is stored in memory space 273, and the E~-:representation on lead 265 is stored in memory space 275.
In response to the aforementioned request from client terminal 130 for tran~;mission of the selected musical piece thereto, processor 280 causes packetizer 285 to generate a stream of packets including one or more of the stored representations of the selected musical piece, depending on the given connection speed. Each packet in the stream is destined for client terminal 13c) as it contains in its header, as a destination address, the IP address of P. Kroon 7-9 15 terminal 130 requesting the music-on-demand service.
Specifically, if the given connection speed is 100 kb/sec, packetizer 285 retrieves from memory spaces 271, 273 and 275 the C-representation, E~-representation and E--representation of the selected musical piece, and packetizes the retrieved representations in accordance with the TCP/IP format. Th.e resulting packet stream is forwarded by processor' 280 to Internet 120.
Fig. 4 illustrates such a packet stream, wherein packets 411, 413 and 415 generated by packetizer 285 respectively contain C'-representation, E;-representation and E~-representation information corresponding to a first time segment of the musical piece; packets 421, 423 and 425 respectively contain C:-representation, E1-representation and E~-representation information corresponding to a second time segment of the musical piece; and so on so forth. To facilitate the assembly of the packets by client terminal 130 when it receives them, the header of each packet contains synchronization information. In particular, the synchronization information in each packet includes a pair of indexes where a sequence index indicating the time segment to which the packet corresponds, followed by a representation index indicating one of the representations with which the packet is associated. For example, field 401 in the header of packet 411 contains the index pair (l, 0), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index '''0" indicating that the packet is associated with the C-representation. Similarly, field 403 in the header of packet 413 contains the index pair (l, 1), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index "1" indicating that the packet is associated with the E;-representation. Field 405 in the header of packet 415 contains the index pair (1, 2), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index "2" indicating that the packet is associated with the E~-representation.
Similarly, the sequence index in each of packets 421, 423 and 425 has a value "2" indicating that the packet corresponds to the second time segment. In addition, the representation indexes of packets 421, 423 and 425 have values "0," "1," and "2", respectively, indicating their respective associations with the C-representation, E1-representation and E~-representation.
Client terminal 130 processes the packet stream from server 105 in accordance with a routine which may be realized using software and/or hardware installed in terminal 130. Fig. 5 illustrates such a routine, denoted 500, where at step 503 terminal 130 receives from server 105 information concerning the indexes identifying the different representations provided thereby to terminal 130.
In this example where the connection speed is 100 kb/sec, as mentioned before terminal 130 is provided with the C-representation, E1-representation and E2-representation of the musical piece which are identified by representation indexes "0," "1," and "2," respectively. Accordingly, upon receipt of the packet stream of Fig. 4, terminal 130 processes the packets on a time segment by time segment basis, and expects to receive three packets associated with the respective representations for each time segment i, 1 i < N, where N is the total number of time segments which the musical piece comprises. In this illustrative embodiment, each time segment has the same predetermined length.
Specifically, at step 507, for each time segment i, terminal 130 sets a predetermined time limit within which any packets associated with the time segment are received P. Kroon 7-9 for processing. Terminal 130 at step 511 examines the aforementioned index pair in the header of each received packet. Based on the sequence index value and the representation index value of the received packets, terminal 130 at step 514 determines whether all of the expected packets for time segment i have been received before the time limit expires. If all of the expected packets have been received, routine 500 proceeds to step 517 where terminal 130 extracts the representation information contents from the respective packets. At step 521, terminal 130 performs on the extracted information the inverse function to embedded audio coder 203 described above to recover a(t) corresponding to time segment i. In particular, in this example where the extracted information includes C-representation information, E1-representation information and E~-representation information, respectively, the aforementioned center decoding is performed on the E~-re:presentation information and E~-representation information based on their correlation to minimize the average distortion in the recovered a(t).
Otherwise, if they aforementioned time limit expires before all of the expected packets are received for time segment i, terminal 13U at step 524 determines whether any received packets for t:he time segment includes the packet containing C-representation information. If it is determined that at least the packet containing C-representation information has been received, terminal 130 extracts representation information contents) from the received packets) for time segment i, and based on the extracted information recovers a(t) corresponding to time segment i, as indicated at step 527. In that case, the audio recovery may be based on only C-representation information corresponding to 28.8_kb/s quality, or on C-representation information in combination with either E~-representation information or E~-representation information corresponding to 64 kb/s quality. Otherwise, if no packet P. Kroon 7-9 1$
containing C-representation information has been received, terminal 130 does not perform any recovery using the received packets for time segment i as any such recovery results in a non-viable a(t). Rather, terminal 130 performs well known audio concealment for time segment i, e.g., interpolation based on the results of audio recovery in neighboring time segments, as indicated at step 531.
If the given connection speed is 64 kb/sec or 28.8 kb/sec instead cf 100 kb/sec in the above example, the above-described proce~>s similarly follows, although in the 64 kb/sec connection :>peed case only C-representation information and E1-representation information or E~-representation information are communicated by server 105 to client terminal 130, and in the 28.8 kb/sec connection speed case only C-representation information is communicated.
The foregoing merely illustrates the principles of the invention. It will thus be appreciated that those skilled in the art will be ab=_e to devise numerous other arrangements which embody the principles of the invention and are thus within it=s spirit and scope.
For example, in anticipation of packet losses because of imperfect network conditions, server 105 in the illustrative emx~odiment may implement path diversity by routing streams of packets containing equivalent amounts of audio information through different paths to the same client terminal, Each packet in each stream corresponds to a different time segment of the audio signal to be recovered. For each 'time segment, the client terminal may ~0 use a packet from any one of the streams corresponding tc the time segment. to recover the audio signal. Thus, despite packet __osses, the quality of the recovered signal is maintained as long as the terminal receives one such packet for each time segment. For instance, to deliver an P. Kroon 7-9 19 audio signal at 64 kb/sec, server 105 may transmit to the client terminal a first stream of packets containing C-representation information, a second stream of packets containing E~,-representation information, and a third stream of packets containing E~-representation information which is equivalent tc E1-representation because of use of complementary quantizers 243 and 253, where the second stream and third stream may be routed through different networks to achieve path diversity.
Similarly, server 105 may implement time diversity by transmitting the streams of packets containing equivalent amounts of audio information one after another through the same network with a predetermined delay.
In addition, based on the disclosure heretofore, it is apparent that a person skilled in the art may generate equivalent C-re~~resent:ations, e.g., C~-representation and C_-representation, using complementary quantizers to achieve path anc./or time diversity of such C-representations.
Further, tr,e mult:i-rate coding technique described above is applicable to communications of not only audio information, but. also information concerning text, graphics, video, etc.
Still further, 1T1 the disclosed embodiment, the inventive mufti--rate coding technique is illustratively applied to a packet scaitched communications system.
However, the inventive technique equally applies to broadcasting sy:>tems :including hybrid in-band on channel (IBOC) AM systems, hybrid IBOC FM systems, satellite ~0 broadcasting sy~~tems, Internet radio systems, TV
broadcasting sy:~tems, etc. -Finally, server 105 is disclosed herein in a form in which various server functions are performed by discrete P. Kroon 7-9 functional blocks. He>wever, any one or more of these functions could equally well be embodied in an arrangement in which the fun.ction;> of any one or more of those blocks or indeed, all cf the functions thereof, are realized, for example, by one or more appropriately programmed processors.
INFORMATION
Field of the Invention The invention relates to systems and methods for communications of a signal containing information, and more particularly to communications systems and methods for coding the signal to generate multiple representations thereof.
Background of the Invention Communications of audio information play an important role in multimedia applications, and Internet applications such as a music-on-demand service, music preview for online compact disk (CDy purchases, etc. To efficiently utilize bandwidth to communicate audio information, a perceptual audio coding (PAC) technique has been developed. For details on the PAC technique, one may refer to U.S. Patent No. 5,285,498 issued February 8, 1994 to Johnston; and U.S.
Patent No. 5,040,217 issued August 13, 1991 to Brandenburg et al.
In accordance with such a PAC technique, each of a succession of time domain blocks of an audio signal representing audio information is coded in the frequency domain. Specifically, the frequency domain representation of each block is divided into coder bands, each of which is individually coded, based on psycho-acoustic criteria, in such a way that the audio information is significantly compressed, thereby requiring a smaller number of bits to represent the audio information than would be the case if the audio information were represented in a more simplistic digital format, such as the PCM format.
For example, in providing the aforementioned music-on-demand service, a server connected to the Internet may store PAC compressed versions of each available musical P. Kroon 7-9 piece to serve client needs. Each version of the musical piece corresponds to a different connection speed at which a client, e.g., a personal computer (PC) having a modem, can afford to communicate over the Internet. The quality, or the lack of distortion, of the version of the musical piece increases with the connection speed corresponding thereto. Thus, for instance, if the server supports (a) a plain old telephone service (POTS) connection speed of about 28.8 kb/sec, (b) an integrated services digital l~ network (ISDN) connection speed of about 64 kb/sec, and (c) a dual ISDN connection speed on the order of 100 kb/sec, three corresponding versions of the musical piece having the respective qualities need to be stored in the server.
However, the storage of musical pieces in this manner is undesirably inefficient and occupies much memory space especially when a large number of musical pieces need to be made available.
In delivering the service to a client at a given connection speed, the server may packetize the 2~ corresponding audio information in the storage, and communicate the resulting packets through a packet switched network, e.g., the Internet. However, in the event that some of the packets are lost in transit because of imperfect network or channel conditions, which is likely, 2~ the quality of the received signal representative of a musical piece would be significantly degraded.
Accordingly, there exists needs for efficiently storing and distributing information at different rates, and effectively maintaining the minimum acceptable quality 3~~ of the received signal despite imperfect network or channel conditions.
Summary of the Invention In accordance with the invention, mufti-rate coding is implemented to generate multiple subrate representations of P. Kroon 7-9 3 a signal containing information, e.g., audio information.
These representations are different from one another and may be delivered at rates lower than or equal to the required delivery rates of the information.
For example, in providing the music-on-demand service described above, at least one of the subrate representations, referred to as a "C-representation," may contain core information delivered at a subrate of 28.8 kb/sec. The other subrate representations, referred to as 1 ~ a "E1-representat:ion" and "E~-representation, each may contain enhancement information delivered at a subrate of 36 kb/sec. Because of the design of the mufti-rate coding in accordance with the invention, in this instance recovery of the signal based on the C-representation alone affords l.~ the minimum acceptable 28.8 kb/sec signal quality; recovery of the signal based on the C-representation in combination with either E~-representation or E~-representation affords a higher 64 kb/s~~c signal quality; and recovery of the signal based on the C-representation in combination with 20 both E;-representation and E~-representation affords the highest 100 kb/s~=c signal quality. Advantageously, the server of the af~~rementioned music-on-demand service needs to store in its memory the subrate representations, i.e., 28.8 kb/sec C-representation, 36 kb/sec E1-representation 25 and 36 kb/sec E~-representation, of each musical piece, in lieu of the 28.8 kb/sec, 64 kb/sec and 100 kb/sec versions thereof as in prior art, to accommodate the different connection speeds and quality requirements, thereby effectively saving the memory space.
30 In accordance with an aspect of the invention, when the subrate representations are communicated to a client terminal in the Form of packets, each packet includes at least an informa~~ion content derived from one of the representations, and an indicator identifying the 3_'~ representation f..om which the information content is derived. Despite losses of some packets in transit because of imperfect channel or network conditions, the signal is recovered based on at least the received packets indicated to contain C-representation information to maintain the minimum acceptable signal quality.
In accordance with one aspect of the present invention there is provided an apparatus for processing a signal comprising: a coder for generating at least first and second representations of the signal, the first and second representations being different from each other; and a controller for packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
In accordance with another aspect of the present invention there is provided a method for processing a 2S signal comprising: generating at least first and second representations of the signal, the first and second representations being different from each other; and packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the 4a information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
Brief Description of the Drawings Fig. 1 illustrates an arrangement embodying the principles of the invention for communicating audio information through a communication network;
Fig. 2 is a block diagram of a server in the arrangement of Fig. 1;
Fig. 3A illustrates a homogeneous multidimensional lattice based on which a prior art quantizer performs quantization;
Fig. 3B illustrates a first non-homogeneous multidimensional lattice based on which a first complementary quantizer in the server of Fig. 2 performs quantization;
Fig. 3C illustrates a second non-homogeneous multidimensional lattice based on which a second complementary quantizer in the server of Fig. 2 performs quantization;
Fig. 4 illustrates a stream of packets generated by the server of Fig. 2; and Fig. 5 is a flow chart depicting the steps whereby a client terminal in the arrangement of Fig. 1 processes the packets from the server.
P. Kroon 7-9 5 Detailed Description Fig. 1 illustrates arrangement 100 embodying the principles of the invention for communicating information, e.g., audio information. In this illustrative embodiment, server 105 in arrangement 100 provides a music-on-demand service to client terminals through Internet 120. One such client terminal is numerically denoted 130 which may be a personal computer (PC). As is well known, Internet 120 is a packet switched network for transporting information in packets in accordance with the standard transmission control protocol/Internet protocol (TCP/IP).
Conventional software including browser software, e.g., the NETSCAPE NAVIGATOR or MICROSOFT EXPLORER browser is installed in client terminal 130 for communicating information with server 105, which is identified by a predetermined uniform resource locator (URL) on Internet 120. For example, to request the music-on-demand service provided by server 105, a modem lnot shown) in client terminal 130 is used to first establish communication connection 125 with Internet 120. Depending on the telecommunication facility subscribed by the user of client terminal 130, communication connection 125 may be limited by different connection speeds. For instance, a plain old telephone service (POTS) connection typically affords a 2~ connection speed of about 28.8 kb/sec; an integrated services digital network (ISDN) connection typically affords a connection speed of about 64 kb/sec; and a dual ISDN connection typically affords a connection speed on the order of 100 kb/sec.
After the establishment of communication connection 125, in a conventional manner, client terminal 130 is assigned an IP address for its identification. The user at client 130 may then access the music-on-demand service at the predetermined URL identifying server 105, and request a P. Kroon 7-9 selected musical piece from the service. Such a request includes the IP ;address identifying client terminal 130, and its connecti«n speed.
In prior art, in providing the music-on-demand service, a server needs to store versions of each musical piece corresponding to different connection speeds supported by the server. The audio quality (distortion) of a version of the musical piece increases (decreases) with the corresponding connection speed. Thus, if a prior art server supports three connection speeds, e.g., 28.8 kb/sec, 64 kb/sec and 100 kb/sec, the server needs to store three different versions of each musical piece available having the respective qualities. However, the storage of musical pieces in this manner is undesirably inefficient and occupies much memory space especially when a large number of musical pieces need to be made available. In addition, in delivering the service to a client terminal, the server typically sends the audio information in the form of packets through the Internet. However, in the event that 2~ some of the packets are lost in transit because of imperfect network or channel conditions, which is likely, the quality of the received audio information would be significantly degraded..
In accordance with the invention, mufti-rate audio coding is implemented in server 105 to generate subrate representations of each musical piece to save memory space.
Different combinatior.,s of the subrate representations of a musical piece correspond to different connection speeds, and audio qualities of the musical piece. In general, the more subrate representations are communicated to a client terminal, the higher t:he audio quality of the musical piece recovered at the terminal and, of course, the higher the connection speed required of the terminal. For example, in this illustrative embodiment, three subrate representations are used in server 10_'> to serve each musical piece in P. Kroon 7-9 accordance with the invention. One of the subrate representations represents core audio information contained in the musical piece, and is referred to as a "C-representation." The other two subrate representations represent first and second enhancement audio information contained in the musical piece, and are referred to as "E:-representation" and "E~-representation," respectively.
Because of the design of the mufti-rate coding in accordance with the invention, the audio signals recovered based on the C-representation alone, although viable, afford the minimum acceptable quality version of a musical piece; the audio signals recovered based on the C-representation in combination with either E1-representation or E_-representa~ion afford a relatively high quality version of the musical. piece; the audio signals recovered based on the C-representation in combination with both E,-representation and E_.-representation afford the highest quality version of they musical piece. However, any audio signals recovered based only on the E1-representation and/or E---representations are not viable.
An embedded audio coder in accordance with the invention is used in server 105 to generate the C-representation requiring a bit rate of, say, 28.8 kb/sec for communication thereof; the E,.-representation requiring a bit rate of, say, 3Ei kb/sec; and the E~-representation requiring a bit rate c>f, say, 36 kb/sec as well. These bit rates are selected such that if all of the representations are used, the quality of the recovered musical piece version is close to that of a 100 kb/sec version generated by a conventional non--embbeded audio coder. Similarly, the quality of the recovered musical piece version based on a combination of the C-representation with the E1-representation or E~-representatio_n is close to that of a 64 kb/sec version generated by the conventional non-embedded audio c:oder. Apparently, the quality of the recovered musical piece version based on the C
P. Kroon 7-9 representation alone is the same as that of a 28.8 kb/sec version generated by the conventional non-embedded audio coder. Advantageously, server 105 only needs to store in its memory the 28.8 kb/sec C-representation, 36 kb/sec E:-representation and 36 kb/sec E~-representation of each musical piece, in lieu of the 28.8 kb/sec, 64 kb/sec and 100 kb/sec versions thereof as in prior art, to accommodate different connection speeds (e. g., 28.8 kb/sec, 64 kb/sec and 100 kb/sec), thereby saving the memory space.
The aforementioned embbeded audio coder implementing multi-rate coding in accordance with the invention will now be described. Fig. 2 illustrates one such embbeded audio coder, denoted 203, in server 105. An analog signal a(t) representing a musical piece is fed to embedded audio coder 203 in providing the music-on-demand service. In response to such an analog signal, analog-to-digital (A/D) convertor 205 in coder 203 digitizes a(t) in a conventional manner, providing PCM samples of a(t). These PCM samples are fed to both filterbank 209 and perceptual model processor 211.
2~ Filterbank 209 divides the samples into time domain blocks, and performs a modified discrete cosine transform (MDCT) on each block to provide a frequency domain representation therefor. Such a frequency domain representation is bandlimited by low-pass filter (LPF) 213 2~ to the 0 to 10 kHz frequency range in this instance. The resulting MDCT coefficients are grouped by quantizer 215 according to coder bands for quantization. These coder bands approximate the well known critical bands of the human auditory system, although limited to the 0 to 10 kHz 30 frequency range in this instance. Quantizer 215 quantizes the MDCT coefficients corresponding to a given coder band with the same quantizer stepsize.
Perceptual model processor 211 analyzes the audio signal samples and determines the appropriate level of 3~ quantization (i.e., stepsize) for each coder band. This P. Kroon 7-9 level of quantization is determined based on an assessment of how well the audio signal in a given coder band masks noise. Quantizer 215 generates quantized MDCT coefficients for application to loss-less compressor 219, which in this instance performs a conventional Huffman compression process on the quantiz.ed coefficients, resulting in the aforementioned C-representation on lead 261. The output of compressor 219 is fed back to quantizer 215 through rate-loop processor 225. In a conventional manner, the latter adjusts the output of quantizer 215 to ensure that the bit rate of the C-representation is maintained at its target rate, which in this instance is 28.8 kb/sec.
In this illustrative embodiment, the E1-representation and E~-representation are generated by coder 203 for enhancing the quality of the musical piece which contain spectral information concerning relatively high frequency components of the audio signal, e.g., in the 7 to 20 kHz range. To that end, t:he quantized MDCT coefficients from quantizer 215 are subtracted by subtracter 229 from the MDCT output of filterbank 209. The resulting difference signals are duplicated by duplicator 231, and then bandlimited respectively by band-pass filters (BPFs) 223 and 233 to the 7 to 20 kHz range. Each of quantizers 243 and 253 receives a copy of the filtered difference signals and quantizes th.e received signals according to predetermined stepsizes.
Quantizers 243 and 253 may be scalar quantizers or multidimensional quant:izers, and may comprise a complementary qL:antizer pair. Complementary scalar quantizers are well known in the art, and described, e.g., in V. Vaishampayan, "Design of Multiple Description of Scalar Quantizers," IhEE Transactions on Information Theory, Vol. 39, No. 3, May 1993, pp. 821-834. In general, a pair of complementary scalar quantizers may be defined by the following encoding functions f: and f , respectively:
P. Kroon 7-9 10 .f , ~x~ : ~i ~ - x, _,"'_,1 , and where ~ represents the real axis, ml = 2'1 and m2 = 2~'~, where S1 and S2 repre~~ent the bit rates for quantizers 243 and 253, respectively. As is well known, associated with each of the quantized values xi and y~ for f; and f~, respectively, is a range or partition [x, y) on the real axis such that all the values in this range are quantized to x: or y, .
In prior art, to take advantage of the correlation between x-~ and y; from fl and f~ having a complementary relationship, joint decoding, also known as "center decoding, " on (xi, y-) is performed in a de-quantizer to realize the optimum decoded value z~., such that the resulting distortion or quantization error is minimized.
The center decocting function, , performed in the de-quantizer may be expressed as follows:
Ll x y ' x ~~' ml,~=m2 ~ ~Z ~ m ~,y~ k ,=t.~=I k=1 It should be noted that not all (x;, y~) are valid decodable combinations depending upon the overlap between their associated partitions. Let Q,., Q~ and Q- be the c0 average distortions associated with f:, f~ and center decoding function , respectively, and let's assume that f:
and f_ are equivalent, i . a . , S1 = S2 = S . I f Q~_ < 2--~ and Q_ < 2--', by minimizing Q- subject to the condition Q1 and P. Kroon 7-9 11 Q ~ Q, where Q is a predetermined distortion value, it can be shown that tr.e value of Q- is always greater than the following limit:
Q ~ 1 2_,~
That is, use of the complementary scalar quantizers affords at most a 3 dB gain, compared with the case where only an individual scal~ir quantizer is used.
However, it. has been recognized that the average distortion O associated with center decoding can be improved if the complE=_mentary quantizers used are l.0 multidimensiona__, rat)zer than scalar as in prior art. In this illustrative embodiment, quantizers 243 and 253 are complementary multidimensional quantizers in accordance with the invention. Preferably, they are non-homogeneous multidimensional lattice quantizers.
._5 In order to more appreciate the advantages of use of complementary non-homogeneous multidimensional lattice quantizers in accordance with the invention, let's first consider a prior art homogeneous 2-dimensional lattice quantizer using a square lattice in a 2-dimensional region :?0 for quantizatio:n. Fig. 3A illustrates one such 2-dimensional region which is defined by X1 and X2 axes and denoted 360. Region 360 in this instance has a square lattice and contains Voronoi regions or cells, e.g., cells 367 and 369, whose length is denoted D, where D represents ?5 a predetermined value. As shown in Fig. 3A, these cells are homogeneously distributed throughout region 360, and are each identified by a different code. As is well known, in the quantization process, the prior art quantizer assigns to an input ;ample point (xl, x2) the code 30 identifying the cell in which the sample point falls, where P. Kroon 7-9 12 xl E X1 and x2 E X2. For example, sample points having 0 xl < D, and 0 < x2 < D are each assigned the code identifying cell 367. In addition, sample points having D
xl < 2D, and D ~ x2 < 2D are each assigned the code identifying cell 369. In practice, each code assignment is achieved by looking up a codebook.
The above prior art quantizer imposes an average distortion proportional to D= which in turn is proportional to 2--s, where in the multidimensional case here S
represents the number of bits/sample/dimension multiplied by the sample rate.
As mentioned before, in the preferred embodiment, quantizers 243 and 253 are complementary non-homogeneous multidimensional lattice quantizers. For example, in the 2-dimensional case, quantizers 243 and 253 use non-homogeneous rectangular lattices in 2-dimensional regions 370 and 390, respectively. In Fig. 3B, like region 360, region 370 is defined by X1 and X2 axes. However, unlike region 360, region 370 contains Voronoi regions or cells, e.g., cells 367 and 369, which are in different shapes and thus non-homogeneous throughout region 370. By way of example, the vertical boundaries of the rectangular cells in region 370 intersect the Xl axis at xl = 0, 0.5~, 2.0~, 2.5~, 4.0~ ..., with the separations between successive vertical boundaries alternating between 0.5~ and 1.5~. On the other hand, the horizontal boundaries of the rectangular cells in region 370 intersect the X2 axis at x2 - 0, 1.5~, 2.0~, 3.5~, 4.00 ..., with the separations between successive horizontal boundaries alternating between 1.5~ and 0.50. In the quantization process, quantizer 343 assigns to an input sample point (xl, x2) the code identifying the cell in which the sample point falls.
For example, sample paints having 0 < xl < 0.5~, and 0 x2 < 1.5~ are each as~.igned the code identifying cell 377.
P. Kroon 7-9 13 In addition, sample points having 0.5~ < xl < 2.0~, and 1.5~ < x2 < 2.04 are each assigned the code identifying cell 379.
A simple way of designing the rectangular lattice in region 390 of quantizer 253, which is complementary to quantizer 243, is to adopt the vertical and horizontal boundaries in region 370 as the horizontal and vertical boundaries in region 390, respectively. Fig. 3C
illustrates the resulting region 390 containing cells, e.g., cells 391 and 399, which are in different shapes, and thus non-homogeneous throughout region 390. In the quantization process, quantizer 253 assigns to an input sample point (xl, x2) the code identifying the cell in which the sample point falls. For example, sample points having 0 < xl < 1.5~, and 0 ~ x2 < 0.5~ are each assigned the code identifying cell 397. In addition, sample points having 1.5~ < xl < 2.00, and 0.5~ < x2 < 2.0~ are each assigned the code identifying cell 399.
It can be shown that the average distortion for an 2J individual one of quantizers 243 and 253 equals 1.25 a 2--', where a represents a constant which depends on the probability density function of the input signal to the quantizer, and S in this instance equals 36 kb/s. However, stemming from the fact that quantizers 243 and 253 are 2~ complementary quantizers, center decoding on the quantized values from quantizers 243 and 253 respectively can be performed in a de-quantizer. It can be shown that the resulting average distortion Q- associated with 2-dimensional center decoding is no more than 0.25 a 2-='.
3~~ That is, complementary quantizers 243 and 253 when implemented with the 2-dimensional center decoding command a 6 dB improvement in terms of distortion over their scalar counterparts.
The equivalent lattices of three and higher dimensions P. Kroon 7-9 14 of complementary quantizers may be obtained similarly to those of two dimensions described above. However, in three or higher dimensions, it is more advantageous to use a non-homogeneous, non-rectangular (or non-hypercube) lattice in each complementary quantizer.
Referring back to Fig. 2, the quantized signals from quantizer 243 are fed to loss-less compressor 245 which, like compressor 219, achieves bit compression on the quantized signals, resulting in the E;-representation on lead 263. The E,-representation is fed back to quantizer 243 through rate-loop processor 247 to ensure that the bit rate of the E1-representation is maintained at its target rate, which in this instance is S1 = 36 kb/sec.
Similarly, the qu.antized signals from quantizer 253 are fed to loss-less compressor 255 which achieves bit compression on the qua.ntized signals, resulting in the E=-representation on lead. 265. The E~-representation is fed back to quantizer 253 through rate-loop processor 257 to ensure that the bit rate of E~-representation is maintained at its target rate, which in this instance is S2 = 36 kb/sec.
Leads 261, 263 and 265 extend to storage 270 where the C-representation on lead 261 is stored in memory space 271, the E1-representation on lead 263 is stored in memory space 273, and the E~-:representation on lead 265 is stored in memory space 275.
In response to the aforementioned request from client terminal 130 for tran~;mission of the selected musical piece thereto, processor 280 causes packetizer 285 to generate a stream of packets including one or more of the stored representations of the selected musical piece, depending on the given connection speed. Each packet in the stream is destined for client terminal 13c) as it contains in its header, as a destination address, the IP address of P. Kroon 7-9 15 terminal 130 requesting the music-on-demand service.
Specifically, if the given connection speed is 100 kb/sec, packetizer 285 retrieves from memory spaces 271, 273 and 275 the C-representation, E~-representation and E--representation of the selected musical piece, and packetizes the retrieved representations in accordance with the TCP/IP format. Th.e resulting packet stream is forwarded by processor' 280 to Internet 120.
Fig. 4 illustrates such a packet stream, wherein packets 411, 413 and 415 generated by packetizer 285 respectively contain C'-representation, E;-representation and E~-representation information corresponding to a first time segment of the musical piece; packets 421, 423 and 425 respectively contain C:-representation, E1-representation and E~-representation information corresponding to a second time segment of the musical piece; and so on so forth. To facilitate the assembly of the packets by client terminal 130 when it receives them, the header of each packet contains synchronization information. In particular, the synchronization information in each packet includes a pair of indexes where a sequence index indicating the time segment to which the packet corresponds, followed by a representation index indicating one of the representations with which the packet is associated. For example, field 401 in the header of packet 411 contains the index pair (l, 0), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index '''0" indicating that the packet is associated with the C-representation. Similarly, field 403 in the header of packet 413 contains the index pair (l, 1), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index "1" indicating that the packet is associated with the E;-representation. Field 405 in the header of packet 415 contains the index pair (1, 2), with the sequence index "1" indicating that the packet corresponds to the first time segment, and the representation index "2" indicating that the packet is associated with the E~-representation.
Similarly, the sequence index in each of packets 421, 423 and 425 has a value "2" indicating that the packet corresponds to the second time segment. In addition, the representation indexes of packets 421, 423 and 425 have values "0," "1," and "2", respectively, indicating their respective associations with the C-representation, E1-representation and E~-representation.
Client terminal 130 processes the packet stream from server 105 in accordance with a routine which may be realized using software and/or hardware installed in terminal 130. Fig. 5 illustrates such a routine, denoted 500, where at step 503 terminal 130 receives from server 105 information concerning the indexes identifying the different representations provided thereby to terminal 130.
In this example where the connection speed is 100 kb/sec, as mentioned before terminal 130 is provided with the C-representation, E1-representation and E2-representation of the musical piece which are identified by representation indexes "0," "1," and "2," respectively. Accordingly, upon receipt of the packet stream of Fig. 4, terminal 130 processes the packets on a time segment by time segment basis, and expects to receive three packets associated with the respective representations for each time segment i, 1 i < N, where N is the total number of time segments which the musical piece comprises. In this illustrative embodiment, each time segment has the same predetermined length.
Specifically, at step 507, for each time segment i, terminal 130 sets a predetermined time limit within which any packets associated with the time segment are received P. Kroon 7-9 for processing. Terminal 130 at step 511 examines the aforementioned index pair in the header of each received packet. Based on the sequence index value and the representation index value of the received packets, terminal 130 at step 514 determines whether all of the expected packets for time segment i have been received before the time limit expires. If all of the expected packets have been received, routine 500 proceeds to step 517 where terminal 130 extracts the representation information contents from the respective packets. At step 521, terminal 130 performs on the extracted information the inverse function to embedded audio coder 203 described above to recover a(t) corresponding to time segment i. In particular, in this example where the extracted information includes C-representation information, E1-representation information and E~-representation information, respectively, the aforementioned center decoding is performed on the E~-re:presentation information and E~-representation information based on their correlation to minimize the average distortion in the recovered a(t).
Otherwise, if they aforementioned time limit expires before all of the expected packets are received for time segment i, terminal 13U at step 524 determines whether any received packets for t:he time segment includes the packet containing C-representation information. If it is determined that at least the packet containing C-representation information has been received, terminal 130 extracts representation information contents) from the received packets) for time segment i, and based on the extracted information recovers a(t) corresponding to time segment i, as indicated at step 527. In that case, the audio recovery may be based on only C-representation information corresponding to 28.8_kb/s quality, or on C-representation information in combination with either E~-representation information or E~-representation information corresponding to 64 kb/s quality. Otherwise, if no packet P. Kroon 7-9 1$
containing C-representation information has been received, terminal 130 does not perform any recovery using the received packets for time segment i as any such recovery results in a non-viable a(t). Rather, terminal 130 performs well known audio concealment for time segment i, e.g., interpolation based on the results of audio recovery in neighboring time segments, as indicated at step 531.
If the given connection speed is 64 kb/sec or 28.8 kb/sec instead cf 100 kb/sec in the above example, the above-described proce~>s similarly follows, although in the 64 kb/sec connection :>peed case only C-representation information and E1-representation information or E~-representation information are communicated by server 105 to client terminal 130, and in the 28.8 kb/sec connection speed case only C-representation information is communicated.
The foregoing merely illustrates the principles of the invention. It will thus be appreciated that those skilled in the art will be ab=_e to devise numerous other arrangements which embody the principles of the invention and are thus within it=s spirit and scope.
For example, in anticipation of packet losses because of imperfect network conditions, server 105 in the illustrative emx~odiment may implement path diversity by routing streams of packets containing equivalent amounts of audio information through different paths to the same client terminal, Each packet in each stream corresponds to a different time segment of the audio signal to be recovered. For each 'time segment, the client terminal may ~0 use a packet from any one of the streams corresponding tc the time segment. to recover the audio signal. Thus, despite packet __osses, the quality of the recovered signal is maintained as long as the terminal receives one such packet for each time segment. For instance, to deliver an P. Kroon 7-9 19 audio signal at 64 kb/sec, server 105 may transmit to the client terminal a first stream of packets containing C-representation information, a second stream of packets containing E~,-representation information, and a third stream of packets containing E~-representation information which is equivalent tc E1-representation because of use of complementary quantizers 243 and 253, where the second stream and third stream may be routed through different networks to achieve path diversity.
Similarly, server 105 may implement time diversity by transmitting the streams of packets containing equivalent amounts of audio information one after another through the same network with a predetermined delay.
In addition, based on the disclosure heretofore, it is apparent that a person skilled in the art may generate equivalent C-re~~resent:ations, e.g., C~-representation and C_-representation, using complementary quantizers to achieve path anc./or time diversity of such C-representations.
Further, tr,e mult:i-rate coding technique described above is applicable to communications of not only audio information, but. also information concerning text, graphics, video, etc.
Still further, 1T1 the disclosed embodiment, the inventive mufti--rate coding technique is illustratively applied to a packet scaitched communications system.
However, the inventive technique equally applies to broadcasting sy:>tems :including hybrid in-band on channel (IBOC) AM systems, hybrid IBOC FM systems, satellite ~0 broadcasting sy~~tems, Internet radio systems, TV
broadcasting sy:~tems, etc. -Finally, server 105 is disclosed herein in a form in which various server functions are performed by discrete P. Kroon 7-9 functional blocks. He>wever, any one or more of these functions could equally well be embodied in an arrangement in which the fun.ction;> of any one or more of those blocks or indeed, all cf the functions thereof, are realized, for example, by one or more appropriately programmed processors.
Claims (44)
1. Apparatus for processing a signal comprising:
a coder for generating at least first and second representations of the signal, the first and second representations being different from each other; and a controller for packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
a coder for generating at least first and second representations of the signal, the first and second representations being different from each other; and a controller for packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
2. The apparatus of claim 1 wherein the plurality of packets include at least first and second sequences of packets, the indicator in each packet in the first sequence identifying the first representation, and the indicator in each packet in the second sequence identifying the second representation.
3. The apparatus of claim 2 wherein the signal is processed on a time-segment basis, each time segment being associated with a different packet in the first sequence and another different packet in the second sequence.
4. The apparatus of claim 3 wherein each packet also includes a second indicator identifying the time segment with which the packet is associated.
5. The apparatus of claim 1 wherein the signal contains audio information.
6. The apparatus of claim 5 wherein the signal is encoded in accordance with a perceptual audio coding (PAC) technique.
7. The apparatus of claim 1 wherein the signal comprises a plurality of frequency components, the first representation being derived from a first subset of the frequency components, and the second representation being derived from a second subset of the frequency components.
8. The apparatus of claim 7 wherein the first subset of the frequency components is identical to the second subset of the frequency components.
9. The apparatus of claim 8 wherein the first representation is derived using a first quantizer, and the second representation is derived using a second quantizer, the first and second quantizers being complementary to each other.
10. The apparatus of claim 7 wherein the first subset of the frequency components is different from the second subset of the frequency components, the first representation providing a description of the signal, the second representation providing enhancement to the description.
11. The apparatus of claim 1 wherein the first and second representations are delivered at different rates.
12. The apparatus of claim 2 wherein the first sequence of packets and the second sequence of packets are provided at different times.
13. The apparatus of claim 1 wherein the first sequence of packets and the second sequence of packets are provided via different communication paths.
14. Apparatus for recovering a signal comprising:
an interface for receiving a plurality of packets, each packet including an indicator, and an information content derived from one of a plurality of representations of the signal, the indicator identifying the representation from which the information content is derived, the plurality of representations being different from one another, wherein the manner in which the representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal; and a processor associated with the client terminal and responsive to the received packets for recovering the signal.
an interface for receiving a plurality of packets, each packet including an indicator, and an information content derived from one of a plurality of representations of the signal, the indicator identifying the representation from which the information content is derived, the plurality of representations being different from one another, wherein the manner in which the representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal; and a processor associated with the client terminal and responsive to the received packets for recovering the signal.
15. The apparatus of claim 14 wherein the interface also receives information concerning the number of representations.
16. The apparatus of claim 14 wherein the plurality of packets include at least first and second sequences of packets, and the plurality of representations include at least first and second representations, the indicator in each packet in the first sequence identifying the first representation, and the indicator in each packet in the second sequence identifying the second representation.
17. The apparatus of claim 16 wherein the signal is recovered on a time-segment basis, each time segment being associated with a different packet in the first sequence and another different packet in the second sequence.
18. The apparatus of claim 17 wherein each packet also includes a second indicator identifying the time segment with which the packet is associated.
19. The apparatus of claim 17 wherein the first representation provides a description of the signal, and the second representation provides enhancement to the description.
20. The apparatus of claim 19 wherein the processor performs concealment for a time segment in recovering the signal when the packet in the first sequence associated with the time segment is not received within a predetermined period.
21. The apparatus of claim 14 wherein the signal contains audio information.
22. The apparatus of claim 21 wherein the signal is encoded in accordance with a PAC technique.
23. A method for processing a signal comprising:
generating at least first and second representations of the signal, the first and second representations being different from each other; and packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
generating at least first and second representations of the signal, the first and second representations being different from each other; and packaging at least one of the first and second representations into a plurality of packets, each packet including at least an indicator, and an information content derived from one of the first and second representations, the indicator identifying the representation from which the information content is derived, wherein the manner in which the one or more representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal.
24. The method of claim 23 wherein the plurality of packets include at least first and second sequences of packets, the indicator in each packet in the first sequence identifying the first representation, and the indicator in each packet in the second sequence identifying the second representation.
25. The method of claim 24 wherein the signal is processed on a time-segment basis, each time segment being associated with a different packet in the first sequence and another different packet in the second sequence.
26. The method of claim 25 wherein each packet also includes a second indicator identifying the time segment with which the packet is associated.
27. The method of claim 23 wherein the signal contains audio information.
28. The method of claim 27 wherein the signal is encoded in accordance with a PAC technique.
29. The method of claim 23 wherein the signal comprises a plurality of frequency components, the first representation being derived from a first subset of the frequency components, and the second representation being derived from a second subset of the frequency components.
30. The method of claim 29 wherein the first subset of the frequency components is identical to the second subset of the frequency components.
31. The method of claim 30 wherein the first representation is derived using a first quantizer, and the second representation is derived using a second quantizer, the first and second quantizers being complementary to each other.
32. The method of claim 29 wherein the first subset of the frequency components is different from the second subset of the frequency components, the first representation providing a description of the signal, the second representation providing enhancement to the description.
33. The method of claim 23 wherein the first and second representations are delivered at different rates.
34. The method of claim 24 wherein the first sequence of packets and the second sequence of packets are provided at different times.
35. The method of claim 24 wherein the first sequence of packets and the second sequence of packets are provided via different communication paths.
36. A method for recovering a signal comprising:
receiving a plurality of packets, each packet including an indicator, and an information content derived from one of a plurality of representations of the signal, the indicator identifying the representation from which the information content is derived, the plurality of representations being different from one another, wherein the manner in which the representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal; and recovering the signal in response to the received packets.
receiving a plurality of packets, each packet including an indicator, and an information content derived from one of a plurality of representations of the signal, the indicator identifying the representation from which the information content is derived, the plurality of representations being different from one another, wherein the manner in which the representations are packaged into the plurality of packets for delivery to a given client terminal is based at least in part on a connection speed associated with a connection over which the plurality of packets are to be delivered to the client terminal; and recovering the signal in response to the received packets.
37. The method of claim 36 further comprising receiving information concerning the number of representations.
38. The method of claim 36 wherein the plurality of packets include at least first and second sequences of packets, and the plurality of representations include at least first and second representations, the indicator in each packet in the first sequence identifying the first representation, and the indicator in each packet in the second sequence identifying the second representation.
39. The method of claim 38 wherein the signal is recovered on a time-segment basis, each time segment being associated with a different packet in the first sequence and another different packet in the second sequence.
40. The method of claim 39 wherein each packet also includes a second indicator identifying the time segment with which the packet is associated.
41. The method of claim 39 wherein the first representation provides a description of the signal, and the second representation provides enhancement to the description.
42. The method of claim 41 further comprising performing concealment for a time segment in recovering the signal when the packet in the first sequence associated with the time segment is not received within a predetermined period.
43. The method of claim 36 wherein the signal contains audio information.
44. The method of claim 43 wherein the signal is encoded in accordance with a PAC technique.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US09/280,785 | 1999-03-29 | ||
| US09/280,785 US6366888B1 (en) | 1999-03-29 | 1999-03-29 | Technique for multi-rate coding of a signal containing information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2300495A1 CA2300495A1 (en) | 2000-09-29 |
| CA2300495C true CA2300495C (en) | 2005-01-11 |
Family
ID=23074644
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002300495A Expired - Fee Related CA2300495C (en) | 1999-03-29 | 2000-03-13 | Technique for multi-rate coding of a signal containing information |
Country Status (5)
| Country | Link |
|---|---|
| US (2) | US6366888B1 (en) |
| EP (2) | EP1575033B1 (en) |
| JP (1) | JP4004707B2 (en) |
| CA (1) | CA2300495C (en) |
| DE (1) | DE60045135D1 (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6556666B1 (en) * | 1998-05-05 | 2003-04-29 | Siemens Information & Communication Networks, Inc. | Notification system for multimedia messaging systems |
| US6366888B1 (en) * | 1999-03-29 | 2002-04-02 | Lucent Technologies Inc. | Technique for multi-rate coding of a signal containing information |
| US6930983B2 (en) * | 2000-03-15 | 2005-08-16 | Texas Instruments Incorporated | Integrated circuits, systems, apparatus, packets and processes utilizing path diversity for media over packet applications |
| US7606164B2 (en) * | 1999-12-14 | 2009-10-20 | Texas Instruments Incorporated | Process of increasing source rate on acceptable side of threshold |
| JP3507743B2 (en) * | 1999-12-22 | 2004-03-15 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Digital watermarking method and system for compressed audio data |
| GB2367219A (en) * | 2000-09-20 | 2002-03-27 | Vintage Global | Streaming of media file data over a dynamically variable bandwidth channel |
| EP1215663A1 (en) * | 2000-12-15 | 2002-06-19 | BRITISH TELECOMMUNICATIONS public limited company | Encoding audio signals |
| US8776153B2 (en) * | 2001-04-25 | 2014-07-08 | At&T Intellectual Property I, Lp | Method and system for transferring content to a networked unit |
| US20030009578A1 (en) * | 2001-07-03 | 2003-01-09 | Apostolopoulos John G. | Method and system for delivering streaming media to fixed clients or mobile clients with multiple description bitstreams |
| US20080030623A1 (en) * | 2001-07-19 | 2008-02-07 | Kumar Ramaswamy | Robust reception of digital broadcast transmission |
| US20030236674A1 (en) * | 2002-06-19 | 2003-12-25 | Henry Raymond C. | Methods and systems for compression of stored audio |
| GB2392807A (en) * | 2002-09-06 | 2004-03-10 | Sony Uk Ltd | Processing digital data |
| US8027381B2 (en) * | 2003-01-28 | 2011-09-27 | Thomson Licensing | Robust mode staggercasting user controlled switching modes |
| US7810124B2 (en) * | 2003-01-28 | 2010-10-05 | Thomson Licensing | Robust mode staggercasting fast channel change |
| ES2353872T3 (en) * | 2003-01-28 | 2011-03-07 | Thomson Licensing | DIFFUSION SPACED IN ROBUST MODE. |
| US7398207B2 (en) * | 2003-08-25 | 2008-07-08 | Time Warner Interactive Video Group, Inc. | Methods and systems for determining audio loudness levels in programming |
| PL1839297T3 (en) | 2005-01-11 | 2019-05-31 | Koninklijke Philips Nv | Scalable encoding/decoding of audio signals |
| JP4725127B2 (en) * | 2005-02-16 | 2011-07-13 | ソニー株式会社 | Decoding apparatus and method, recording medium, and program |
| US8620644B2 (en) * | 2005-10-26 | 2013-12-31 | Qualcomm Incorporated | Encoder-assisted frame loss concealment techniques for audio coding |
| JP2010538534A (en) * | 2007-08-28 | 2010-12-09 | トムソン ライセンシング | Stagger casting without channel change delay |
| US20100321208A1 (en) * | 2009-06-23 | 2010-12-23 | Craig Stephen Etchegoyen | System and Method for Emergency Communications |
| KR101786050B1 (en) * | 2009-11-13 | 2017-10-16 | 삼성전자 주식회사 | Method and apparatus for transmitting and receiving of data |
| WO2013113351A1 (en) * | 2012-01-30 | 2013-08-08 | Nokia Siemens Networks Oy | Processing a signal in station apparatus |
| US8595590B1 (en) | 2012-12-03 | 2013-11-26 | Digital PowerRadio, LLC | Systems and methods for encoding and decoding of check-irregular non-systematic IRA codes |
| US9191256B2 (en) | 2012-12-03 | 2015-11-17 | Digital PowerRadio, LLC | Systems and methods for advanced iterative decoding and channel estimation of concatenated coding systems |
| US8948272B2 (en) | 2012-12-03 | 2015-02-03 | Digital PowerRadio, LLC | Joint source-channel decoding with source sequence augmentation |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| NL9000338A (en) * | 1989-06-02 | 1991-01-02 | Koninkl Philips Electronics Nv | DIGITAL TRANSMISSION SYSTEM, TRANSMITTER AND RECEIVER FOR USE IN THE TRANSMISSION SYSTEM AND RECORD CARRIED OUT WITH THE TRANSMITTER IN THE FORM OF A RECORDING DEVICE. |
| US6047112A (en) * | 1992-06-30 | 2000-04-04 | Discovision Associates | Technique for initiating processing of a data stream of encoded video information |
| TW272341B (en) * | 1993-07-16 | 1996-03-11 | Sony Co Ltd | |
| US5881104A (en) * | 1996-03-25 | 1999-03-09 | Sony Corporation | Voice messaging system having user-selectable data compression modes |
| US6154499A (en) * | 1996-10-21 | 2000-11-28 | Comsat Corporation | Communication systems using nested coder and compatible channel coding |
| US6366888B1 (en) * | 1999-03-29 | 2002-04-02 | Lucent Technologies Inc. | Technique for multi-rate coding of a signal containing information |
| US6430401B1 (en) * | 1999-03-29 | 2002-08-06 | Lucent Technologies Inc. | Technique for effectively communicating multiple digital representations of a signal |
| US6246345B1 (en) * | 1999-04-16 | 2001-06-12 | Dolby Laboratories Licensing Corporation | Using gain-adaptive quantization and non-uniform symbol lengths for improved audio coding |
-
1999
- 1999-03-29 US US09/280,785 patent/US6366888B1/en not_active Expired - Lifetime
-
2000
- 2000-03-13 CA CA002300495A patent/CA2300495C/en not_active Expired - Fee Related
- 2000-03-22 EP EP05076403A patent/EP1575033B1/en not_active Expired - Lifetime
- 2000-03-22 EP EP00302320.7A patent/EP1049074B1/en not_active Expired - Lifetime
- 2000-03-22 DE DE60045135T patent/DE60045135D1/en not_active Expired - Lifetime
- 2000-03-29 JP JP2000090014A patent/JP4004707B2/en not_active Expired - Fee Related
-
2001
- 2001-11-07 US US10/039,458 patent/US6920422B2/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| EP1575033B1 (en) | 2010-10-20 |
| EP1049074B1 (en) | 2014-06-18 |
| JP4004707B2 (en) | 2007-11-07 |
| US6920422B2 (en) | 2005-07-19 |
| CA2300495A1 (en) | 2000-09-29 |
| US20020091531A1 (en) | 2002-07-11 |
| DE60045135D1 (en) | 2010-12-02 |
| EP1575033A1 (en) | 2005-09-14 |
| US6366888B1 (en) | 2002-04-02 |
| JP2000311000A (en) | 2000-11-07 |
| EP1049074A1 (en) | 2000-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2300495C (en) | Technique for multi-rate coding of a signal containing information | |
| US6044089A (en) | System and method for scaleable audio transmission over a network | |
| US7668723B2 (en) | Scalable lossless audio codec and authoring tool | |
| CA2326495C (en) | Technique for parametric coding of a signal containing information | |
| CA2327627C (en) | Process for processing at least one coded binary audio flux organized into frames | |
| CA1245780A (en) | Method of reconstructing lost data in a digital voice transmission system and transmission system using said method | |
| CN1973320B (en) | Stereo coding and decoding methods and apparatuses thereof | |
| US5570363A (en) | Transform based scalable audio compression algorithms and low cost audio multi-point conferencing systems | |
| US8175729B2 (en) | Preserving matrix surround information in encoded audio/video system and method | |
| US6356545B1 (en) | Internet telephone system with dynamically varying codec | |
| EP2360682B1 (en) | Audio packet loss concealment by transform interpolation | |
| US7617097B2 (en) | Scalable lossless audio coding/decoding apparatus and method | |
| US8374858B2 (en) | Scalable lossless audio codec and authoring tool | |
| US6970479B2 (en) | Encoding and decoding of a digital signal | |
| EP1949693B1 (en) | Method and apparatus for processing/transmitting bit-stream, and method and apparatus for receiving/processing bit-stream | |
| EP2228791B1 (en) | Scalable lossless audio codec and authoring tool | |
| EP1290835A1 (en) | Transmission over packet switched networks | |
| Feerick | How much delay is too much delay? | |
| Isenburg | Transmission of multimedia data over lossy networks | |
| JPS6251827A (en) | Voice coding system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |
Effective date: 20160314 |