WO2016015286A1

WO2016015286A1 - Methods and apparatuses for data compression and decompression

Info

Publication number: WO2016015286A1
Application number: PCT/CN2014/083428
Authority: WO
Inventors: Ang FENG; Chen Wang; Jinsong Yang
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2014-07-31
Filing date: 2014-07-31
Publication date: 2016-02-04
Anticipated expiration: 2017-01-31

Abstract

Methods and apparatuses for IQ data compression and decompression. In one embodiment, the method for IQ data compression comprises receiving an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence has a length of N_dbits. The method further comprises generating a compressed IQ data block from the uncompressed IQ data block. In the generating step, a sign bit segment of the compressed IQ data block is generated by using one or more sign bits of the one or more uncompressed samples, and a set index segment of the compressed IQ data block is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples.

Description

METHODS AND APPARATUSES FOR DATA COMPRESSION AND

DECOMPRESSION

TECHNICAL FIELD

The embodiments of the present disclosure generally relate to data processing, particularly to methods and apparatuses for in-phase and quadrature (IQ) data compression/decompression.

DESCRIPTION OF THE RELATED ART

Growing capacity demand of cellular communication system necessitates a distributed deployment of Radio Unit (RU) such as Remote Radio Heard (RRH) , Radio Equipment (RE) and centralized deployment of Digital Unit such as Baseband Unit (BBU) , Radio Equipment Control (REC) , connected between each other via a transport link carrying in-phase and quadrature (IQ) data. The link capacity becomes bottleneck as the number of antenna or the bandwidth of signal increases.

One solution is employing multiple modules or higher speed module, which makes the cost expensive. Recently, methods for IQ data compression/decompression were proposed to solve the dilemma of low cost and high capacity.

Time-domain compression/decompression is a category of the existing technologies on IQ data compression/decompression. In the time-domain compression/decompression, the data over a transport link are still time-domain samples, but with less redundancy between successive samples and within each sample. The examples of the time-domain methods include source coding, quantization, Automatic Gain Control (AGC) , re-sampling.

With respect to the source coding method, it typically applies Huffman codes to each sample straightforwardly. The source coding method usually uses shorter codeword for higher possible sample, which however may leads to the compression rate uncontrollable.

With respect to the quantization method, it may utilize Lloyd-Max algorithm to obtain a non-uniform partition and codebook, having minimum quantization error. However, the uniform quantization method leads to Error Vector Magnitude (EVM) deterioration, as a consequence, degrades SNR or BER performance . Non-uniform quantization although has superior performance than uniform quantization, it has very high complexity because of using Lloyd-Max algorithm. With respect to the AGC method, also known as scaling, it collects a block samples and extracts their common magnitude, resulting in a low dynamic range and is generally used in addition to quantizer as an option. Furthermore, AGC have to be performed within a block of data, as a result, the processing delay is relative to the block size.

With respect to the re-sampling method, it exploits the fact that part of bandwidth is reserved as guard band. For this reason, down-sample can be used to shrink the guard band, still conforming to Nyquist theory. However, resample will cause performance loss even if the down-sample rate is still larger than the Nyquist bandwidth. Apparently, the present data compression methods can hardly satisfy the practice requirements of the transport link between a RU and a DU due to the unacceptable error rate, performance, delay, and so on. Even if the performance and/or delay are improved, the implementation complexity will become unbearable instead.

Although the problems in the existing IQ data compression/decompression technologies are described in the context of cellular communication system, those skilled in the art can appreciate that similar problems also exist in other kinds of communication system where IQ data are carried by links between communication elements.

SUMMARY OF THE DISCLOSURE

Therefore, there is a need to provide solutions for IQ data compression and decompression that have relatively low complexity but can provide a relatively high compression performance . In order to solve at least one of the above problems in the prior art, one or more method and apparatus embodiments according to the present disclosure aim to provide solutions for IQ data compression and decompression.

Various aspects of examples of the disclosure are set out in the claims.

According to an aspect of the present disclosure, an embodiment of the present disclosure provides a method for IQ data compression. The method comprises receiving an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence has a length of N_dbits . The method further comprises generating a compressed IQ data block from the uncompressed IQ data block. In the generating step, a sign bit segment of the compressed IQ data block is generated by using one or more sign bits of the one or more uncompressed samples, and a set index segment of the compressed IQ data block is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples. Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to another aspect of the present disclosure, an embodiment of the present disclosure provides a method for IQ data decompression. The method comprises receiving a compressed IQ data block including a sign bit segment and a set index segment. The compressed IQ data block can correspond to an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence has a length of N_dbits. The method further comprises recovering the uncompressed IQ data block from the compressed IQ data block. The recovering step comprises recovering one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block, and recovering one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, which is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples. Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides a compressor for IQ data compression. The compressor comprises a receiving unit configured to receive an uncompressed IQ data block including one or more uncompressed samples . Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence has a length of ¾ bits . The compressor further comprises a generating unit configured to generating a compressed IQ data block from the uncompressed IQ data block. The generating unit comprises a sign bit segment generating unit configured to generate a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples, and a set index segment generating unit configured to generate a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples. Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides a decompressor for IQ data decompression. The decompressor comprises a receiving unit configured to receive a compressed IQ data block including a sign bit segment and a set index segment. The compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence has a length of N_dbits. The decompressor further comprises a recovering unit configured to recover the uncompressed IQ data block from the compressed IQ data block, the recovering unit comprises a sign bit recovering unit configured to recover one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block; and a first bit sequence recovering unit configured to recover one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, which is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples . Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides an apparatus for IQ data compression. The apparatus comprises one or more processors; and one or more memories including computer program code . The one or more memories and the computer program code are configured to, with the one or more processors, cause the apparatus to perform the method for IQ data compression according one or more embodiments of the present disclosure.

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides an apparatus for IQ data decompression, comprising one or more processors; and one or more memories including computer program code . The one or more memories and the computer program code are configured to, with the one or more processors, cause the apparatus to perform the decompression method according to one or more embodiments of the present disclosure .

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides a compressing device for IQ data compression. The compressing device comprising processing means adapted to receive an uncompressed IQ data block including one or more uncompressed samples . Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_d bits . The processing means is further adapted to generate a compressed IQ data block from the uncompressed IQ data block by generating a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples, and generating a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples. Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to other aspects of the present disclosure, an embodiment of the present disclosure also provides a decompressing device for IQ data decompression. The decompressing device comprises processing means adapted to receive a compressed IQ data block including a sign bit segment and a set index segment, the compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits. The processing means is further adapted to recover the uncompressed IQ data block from the compressed IQ data block by recovering one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block, and recovering one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, which is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples . Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. Union of a plurality of bit sets constitutes a universal set of N_d-bit sequences. Each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set. According to one or more embodiments of the present disclosure, an uncompressed IQ data block including one or more uncompressed samples can be compressed efficiently into a corresponding compressed IQ data block by separating each uncompressed sample at least into sign bit, set index code, so as to generate a compressed IQ data block with a constant compression rate. The various embodiments of the present disclosure provide a good performance with lower implementation complexity. In addition, based on some embodiments of the present disclosure, compression delay can be controlled to satisfy system requirement.

BRIEF DESCRIPTION OF THE DRAWINGS

Inventive features regarded as the characteristics of the present disclosure are set forth in the appended claims . However, the present disclosure, its implementation mode, other objectives, features and advantages will be better understood through reading the following detailed description on the exemplary embodiments with reference to the accompanying drawings, where in the drawings:

Fig. 1 schematically illustrates an exemplary flow diagram of the method for IQ data compression according to one or more embodiments of the present disclosure;

Fig. 2 schematically illustrates an example of bit set partition and set indices according to one embodiment of the present disclosure;

Fig. 3 schematically illustrates bit mapping between an uncompressed sample and a corresponding compressed sample according to one or more embodiments of the present disclosure ; Fig. 4 schematically illustrates bit mapping between a block of more than one uncompressed samples and a block of corresponding compressed samples, according to one or more embodiments of the present disclosure;

Fig.5 schematically illustrates an exemplary flow diagram of the method for IQ data decompression according to one or more embodiments of the present disclosure;

Fig. 6 schematically illustrates a compressor for IQ data compression according to one or more embodiments of the present disclosure; Fig. 7 schematically illustrates a decompressor for IQ data decompression according to one or more embodiments of the present disclosure; and

Fig. 8 schematically illustrates a device in which one or more embodiments of the present disclosure can be implemented.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

Hereinafter, embodiments of the present disclosure be described with reference to the accompanying drawings . In the following description, many specific details are illustrated so as to understand the present disclosure more comprehensively. However, it is apparent to the skilled in the art that implementation of the present disclosure may not have these details . Additionally, it should be understood that the present disclosure is not limited to the particular embodiments as introduced here . On the contrary, any arbitrary combination of the following features and elements may be considered to implement and practice the present disclosure, regardless of whether they involve different embodiments. Thus, the following aspects, features, embodiments and advantages are only for illustrative purposes, and should not be understood as elements or limitations of the appended claims, unless otherwise explicitly specified in the claims.

Fig. 1 schematically illustrates an exemplary flow diagram of the method 100 for IQ data compression according to one or more embodiments of the present disclosure.

As shown in Fig. 1, in step S110, an uncompressed IQ data block is received, which includes one or more uncompressed samples. Each of the one or more uncompressed samples includes at least one sign bit and a bit sequence, where the bit sequence has a length of iV_dbits.

According to the embodiments of the present disclosure, a predefined number of uncompressed IQ data sample (s) may constitute a block of uncompressed IQ data that will be compressed as a corresponding compressed IQ data block. The size of the block may be selected according to delay requirements. In principle, for delay-sensitive data, the predefined number may be selected as a relatively small value, such as one or two, so as to reduce the delay in decompression. On the other hand, for delay-tolerant data, the predefined number may be selected as a relatively big value, so as to improve the compression performance.

Here, the bit configuration of an uncompressed IQ data sample will be explained by taking a 15 -bit uncompressed IQ data sample as an example. A 15 -bit uncompressed IQ data sample may have only one sign bit (i.e., the length S of the sign bit is "1") and a bit sequence with length of 14 bits (i.e., the length N_d of the bit sequence is "14") . The embodiments of the present disclosure are also adapted to handle any other suitable IQ data sample configuration, without limiting to a specific length of the sign bit or a specific length of the bit sequence.

In order to facilitate compression of bit sequences, a universal set of N_d-bit sequences can be divided into a plurality of bit sets. Each bit sequence can belong to a specific bit set and each element (i.e., bit sequence) belonging to the same bit set has its first portion with a bit pattern that serves as a set index of the bit set. The bit set can be uniquely identified by the set index. Apparently, the set index may have a length equal to or less than N_d, since the set index can be regarded as a portion of the bit sequence elements in the corresponding bit set. According to one or more embodiments of the present disclosure, the length of a set index of a specific bit set may depend upon element number of the specific bit set.

Fig. 2 schematically illustrates an example of bit set partition and set indices according to one embodiment of the present disclosure. In Fig. 2, "1" denotes a bit "1"; "0" denotes a bit "0"; and "X" denotes an arbitrary bit. As shown in Fig. 2, a universal set of 14 -bit sequences may be divided into the number 15 of bit sets, i.e., setO, setl,..., setl4. The set indices are shown in the triangular. For example, for setO, the set index is "1" and the number of bit sequence elements of setO is 2¹⁴; for set3, the set index is "0001" and the number of bit sequence elements in set3 is 2¹⁰. In this example, with the length of the bit set index increasing, the number of elements in the bit set is decreased. Such set indices are efficient for uniform distributive data, when directly used to compress IQ data. However, it is not the case in most of traffic scenarios, as the distribution of bit sequences of the IQ data samples is usually approximate to a normal distribution. As a consequence, the set index represented in Fig. 2 may not be efficient for IQ data compression any more.

According to one or more embodiments of the present disclosure, a set index can be encoded into a corresponding set index code, whose length M₀ is less than the length of corresponding bit sequence N_d. As such, set index code may further reduce the bits that would be used to identify a bit set in IQ data compression and/or to adapt the length M₀ of the set index code to distribution of data. According to one or more embodiments of embodiments of the present disclosure, the length of the set index code M₀ may have a smaller value, if distribution probability of data in the corresponding bit set is higher. Let' s consider the exemplary set partition as illustrated in Fig. 2. The simulation experiments conducted by the Inventors have shown that when IQ data has a distribution approximate to a normal distribution, the distribution having the maximum element number according to the set partition of Fig. 2. Therefore, it would be advantageous to encode the set index of set3 "0001" into the shortest set index code.

For example, the resultant encoding codebook may be given in Table 1 .

Table 1 : Codebook corresponding to possibility distribution of data

In the example of the codebook as shown in Table 1 , the codebook def ines a mapping relationship between set indices and set index codes . The set index codes of corresponding bit sets are selected from the set of prefix codes so that the length of a set index code can ref lect the different possibility distribution of data for the corresponding bit set . Those skilled in the art may appreciate that the set index codes may also be selected from any other suitable type of coding set , as long as it meets relevant encoding and decoding requirements. Therefore, the design of a codebook that can be used to implement an embodiment of the present disclosure will not limit to the example of Table 1, but having various modifications and alternatives. In step S120, a compressed IQ data block is generated from the uncompressed IQ data block. Specifically, a sign bit segment of the compressed IQ data block can be generated, in step S121, by using one or more sign bits of the one or more uncompressed samples, and a set index segment of the compressed IQ data block can be generated, in step S122, by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples.

According to one or more embodiments of the present disclosure, the length of the compressed IQ data block may be configured as a predetermined value, which for example may be a fixed integer, so as to obtain a fixed compression rate.

According to one or more embodiments of the present disclosure, in response to determining that a sum of the length of the sign bit segment and the length of the set index segment is less than the predetermined length for the compressed IQ data block, in step S123, a residual bit segment of the compressed IQ data may be generated by using one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples, such that a sum of the length of the sign bit segment S, the length of the set index segment M₀ and the length of the residual bit segment Ma is equal to the predetermined length for the compressed IQ data block. With reference to Figs. 3 and 4, the step of generating a compressed IQ data block will be described in detail. Fig. 3 schematically illustrates bit mapping between an uncompressed sample and a corresponding compressed sample according to one or more embodiments of the present disclosure . It should be noted that in real IQ data, one IQ sample includes both one in-phase sample and one quadrature sample. From the perspective of compression and/or decompression, the same compression/decompression method can be applied to both in-phase and quadrature paths, without any changes. Therefore, although the example as shown in Fig. 3 only represents one sample, those skilled in the art may understand that such a sample may refer to both the in-phase sample and the quadrature sample .

As shown in Fig. 3, a single uncompressed IQ data sample "000000110011100", denoted by reference numeral 310 includes a 1-bit sign, denoted by 311 and a 14 -bit bit sequence, denoted by 312, i.e., S=l, N_d=14: . The uncompressed sample, denoted by 310, is compressed into an 8-bit compressed sample 320, which includes a sign bit segment, denoted by 321, a set index segment, denoted by 322, and optionally further includes an residual bit segment, denoted by 323.

The sign bit 311 is directly reserved in the compressed sample 320 to generate the sign bit segment 321, which is a bit "0" in this example. According to the set partition as shown in Fig.2, the bit sequence 312 "00000110011100" belongs to set5 with the set index "000001" (i.e., the first portion of the bit sequence 312) . The set index code segment 322 may be obtained by encoding the set index of set5 "000001". For example, according to the codebook specified in Table 1, the set index identifying set3 can be encoded into "011", which is then used to generate the set index code segment 322 of the compressed sample 320. It can be seen that in this example, the sum of the length of the sign bit segment S and the length of the set index segment is less M₀ than the predetermined length for the compressed IQ data block. In response to this, a residual bit segment 323 may be used to ensure the compressed sample having the predetermined length. The remaining 4 bits in the compressed sample 320 is the residual bit segment and the second portion of the bit sequence 312 "1001". As a result, the generated compressed IQ data sample 320 is "00111001". The last four bits "1100" in the uncompressed sample 310 is eventually discarded, which could result in an acceptable compression error. Since the bit length of the compressed IQ data block (sample) , S+M₀+M_d, is a constant configured by system, the compression rate equals to (S+M₀+M_d) / {S+N_d) . In some embodiments of the present disclosure, multiple uncompressed IQ data samples can be used together to generate a compressed IQ data block. The single sample compression can be extended to multi-sample compression. Fig. 4 schematically illustrates bit mapping between a block of more than one uncompressed samples and a block of corresponding compressed samples, according to one or more embodiments of the present disclosure.

As shown in Fig. 4, the number of 2*N_Puncompressed samples, which include N_P in-phase samples and N_P quadrature samples, may form a uncompressed IQ data block, denoted by reference numeral 410, to be compressed into a compressed IQ data block, denoted by 420. The block size can be selected according to the delay requirement. For example, one basic frame of Common Public Radio Interface (CPRI) is about 260ns. If the delay is required to be 260ns, the uncompressed samples within one base frame (BF) may be selected as a block. If the delay is required more than one BF time, e.g. lus, the samples within several BFs may be collected as a block. This mechanism can make sure that the compression delay is always confirmed to the requirement .

Similar with the compression procedure for one single sample described in conjunction with Fig. 3, the sign bits of the samples in the uncompressed IQ data block 410 are used to generate the sign bit segment, denoted by 421" of the compressed IQ data block 420. In the sign bit segment 421, the sign bits of corresponding uncompressed samples may be arranged in the same sequence as the corresponding samples in arranged the received uncompressed IQ data block, or any other predetermined sequence.

The first portions of the corresponding bit sequences of the uncompressed samples, which serve as the set indices of the corresponding bit sequences, are encoded into corresponding set index codes according to a codebook, such as the codebook as illustrated by Table 1. The index codes corresponding to those bit sequences are used to generate the set index segment, denoted by 422, of the compressed IQ data block. The index codes of corresponding uncompressed samples may be arranged in the same sequence as the corresponding samples in arranged the received uncompressed IQ data block, or any other predetermined sequence . One advantage of multi-sample compression is that, the bit set with short encoded index, which occurs at a high possibility, will occupy fewer bits in the compressed IQ data block, which may save the positions for the bit set with long encoded index, which occurs at a low possibility.

In response to determining that the sum of the length of the sign bit segment 2*N_P*S and the length of the set index segment M₀ (n) is less than a predetermined length for the compressed IQ data block, a residual bit segment, denoted by 423, of the compressed IQ data may be generated by using the second portions of corresponding bit sequences of the uncompressed samples, until there is no position in the compressed IQ data block 420.

According to one or more embodiments of the present disclosure, the residual bits in the second portions of the bit sequences will be allocated into the bit positions of the residual bit segment 423 according to a predefined sequence. The sequence may be chosen in a proper way, so as to further improve the compression performance.

In one exemplary implementation, the sequence may be determined by criteria of "Most Significant Bit First", where one bit belongs to an uncompressed sample having the maximum number of residual bits will be firstly allocated into the bit position in the residual bit segment 423. After allocating this bit, if the same uncompressed sample still has the maximum number of residual bits, it continues allocating the next bit in to the residual bit segment 423. Additionally, if an uncompressed sample has the same number of residual bits of with other uncompressed samples, one bit of each equivalent sample will be allocated into the residual bit segment 423 in a predefined sequence, such as the receive sequence . Such allocating is repeated until all bit positions in the compressed IQ data block are occupied.

In another exemplary implementation, the sequence may be determined by criteria of "Most Possible Sample First", where the uncompressed sample having the highest possibility will be allocated firstly into the bit positions in the residual bit segment 423. According to the embodiments described above in conjunction with Fig. 2 and Table 1, based on the length of the corresponding set index code, the uncompressed sample having the highest possibility can be easily identified. All the residual bits of the uncompressed sample with the highest possibility may be allocated into the residual bit segment 423. Subsequently, another uncompressed sample with second highest possibility will be selected and allocated. Repeat such process until all bit positions of the residual bit segment 423 are filled. In case that many uncompressed samples have the same possibility, they may be allocated in a predefined sequence, such as the receive sequence.

After all bit positions of the compressed IQ data block 420 are filled, the remaining residual bits in some uncompressed samples have to be discarded, which then results in an acceptable compression error.

Fig.5 schematically illustrates an exemplary flow diagram of the method 500 for IQ data decompression according to one or more embodiments of the present disclosure. The decompression procedure is contrary to the compression procedure .

As shown in Fig. 5, in step S510, a compressed IQ data block including a sign bit segment and a set index segment is received. The compressed IQ data block corresponds to an uncompressed IQ data block including one or more uncompressed samples, Each of the one or more uncompressed samples includes at least one sign bit and a bit sequence having a length of N_dbits.

In step S520, the uncompressed IQ data block is recovered from the compressed IQ data block. Specifically, one or more sign bits of the one or more uncompressed samples are recovered, in step S521, from the sign bit segment of the compressed IQ data block, and one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples are recovered, in step S522, based on the set index segment of the compressed IQ data block. In the compression procedure, as described above with reference to Fig.2, the set index segment was generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples.

Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs. The length of the set index code M₀ is less than the length of the corresponding bit sequence N_d. The union of a plurality of bit sets constitutes a universal set of N_d-bit sequences .

Each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set. According to one or more embodiments of the present disclosure, the length of a set index of a specific bit set depends upon element number of the specific bit set.

According to one or more embodiments of the present disclosure, the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

According to one or more embodiments of the present disclosure, the one or more set indices of the one or more uncompressed samples are obtained by decoding the one or more set index codes in the set index code segment, according to the same codebook as being used in the compression procedure. The codebook can define a mapping relationship between set indices and set index codes . One example of the codebook is shown in Table 1.

Additionally, the compressed IQ data block may further include a residual bit segment. The recovering step S520 may- further comprises step S523. In step S523, one or more second portions of corresponding one or more sequences of the one or more uncompressed samples may be recovered from the residual bit segment of the compressed IQ data. By decoding the set index codes, the boundary between the first portion (bits of the set index) and the residual bits could be computed. The residual bits are able to be extracted from bits after the boundary. According to the embodiments where multi-sample compression/decompression is implemented, in step S523, the residual bits in the second portions of multiple uncompressed samples may be recovered according to the predefined sequence, which was used to generate the residual bit segment in the compression procedure.

According to one or more embodiments of the present disclosure, if it is determined that an overall length of the first and second portions of the bit sequence of a recovered sample is less than N_d, the recovered sample may be filled (not shown) with bit "0" or bit "1".

It should be noted that the above depiction is only exemplary, not intended for limiting the present disclosure. In other embodiments of the present disclosure, this method may have more, or less, or different steps, and the steps numbering is only for making the depiction more concise and clearer, but not for stringently limiting the sequence between each steps; while the sequence of steps may be different from the depiction. For example, in some embodiments, the above one or more optional steps may be omitted. Specific embodiment of each step may be different from the depiction. All these variations fall within the spirit and scope of the present disclosure . Fig. 6 schematically illustrates a compressor 600 for IQ data compression according to one or more embodiments of the present disclosure.

As shown in Fig. 6, the compressor 600 for IQ data compression comprises a receiving unit 610 and a generating unit 620.

The receiving unit 610 is configured to receive an uncompressed IQ data block including one or more uncompressed samples. Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit and the bit sequence have a length of N_d bits .

The generating unit 620 is configured to generate a compressed IQ data block from the uncompressed IQ data block. The generating unit 620 comprises a sign bit segment generating unit 621 configured to generate a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples , and a set index segment generating unit 622 configured to generate a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples .

Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and the length of the set index code M₀ is less than the length of the corresponding bit sequence N_d.

The union of a plurality of bit sets constitutes a universal set of JW_d-bit sequences. Each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to one or more embodiments of the present disclosure, the length of a set index of a specific bit set may depend upon element number of the specific bit set.

According to one or more embodiments of the present disclosure, the length of the set index code M₀ may have a smaller value, if distribution probability of data in the corresponding bit set is higher.

According to one or more embodiments of the present disclosure, the one or more set index codes may be obtained by encoding one or more set indices of the one or more uncompressed samples according to a codebook, which defines a mapping relationship between set indices and set index codes .

According to one or more embodiments of the present disclosure, the generating unit 620 may further comprise a residual bit segment generating unit 623. The residual bit segment generating unit 623 may be configured to generate, in response to determining that the sum of the length of the sign bit segment and the length of the set index segment is less than a predetermined length for the compressed IQ data block, a residual bit segment of the compressed IQ data by using one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples, such that a sum of the length of the sign bit segment, the length of the set index segment and the length of the residual bit segment is equal to the predetermined length for the compressed IQ data block.

According to one or more embodiments of the present disclosure, the residual bit segment generating unit 623 may be configured to generate a residual bit segment of the compressed IQ data by using bits of bit sequences from the uncompressed samples in a predefined sequence. Fig. 7 schematically illustrates a decompressor 700 for IQ data decompression according to one or more embodiments of the present disclosure.

As shown in Fig. 7, the decompressor 700 comprises a receiving unit 710 and a recovering unit 720.

The receiving unit 710 is configured to receive a compressed IQ data block including a sign bit segment and a set index segment, the compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples . Each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits.

The recovering unit 720 is configured to recover the uncompressed IQ data block from the compressed IQ data block. The recovering unit 720 comprises a sign bit recovering unit 721 configured to recover one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block and a first bit sequence recovering unit 722 configured to recover one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block. The set index segment was generated in a corresponding compressor by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples.

Each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs. The length of the set index code M₀ is less than the length of the corresponding bit sequence ¾· The union of a plurality of bit sets constitutes a universal set of Na^_bit sequences. Each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

According to one or more embodiments of the present disclosure, one or more set indices of the one or more uncompressed samples may be obtained by decoding the one or more set index codes according to a codebook, which defines a mapping relationship between set indices and set index codes .

According to one or more embodiments of the present disclosure, the compressed IQ data block may further include a residual bit segment, and the recovering unit 720 may further comprises a second bit sequence recovering unit 723. The a second bit sequence recovering unit 723 may be configured to recover one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples from a residual bit segment of the compressed IQ data. According to one or more embodiments of the present disclosure, the recovering unit 720 may be further configured to fill a recovered sample with bit "0" or bit "1", in response to determining that an overall length of the first and second portions of the bit sequence of the recovered sample is less than N_d.

According to one or more embodiments of the present disclosure, the device 800 may be a DU or RU or any other suitable device in which one or more embodiments of the present disclosure can be implemented.

The device 800 may be includes a data processor (DP) 810, a memory (MEM) 820 coupled to/embedded in the DP 810. The MEM 820 stores a program (PROG) 830. When the device 800 is operating as a device for compressing IQ data, the PROG 830 is assumed to include program instructions that, when executed by the DP 810, enable the device 800 to operate in accordance with the exemplary embodiments of this disclosure, as discussed herein with the flow diagram as shown in Fig.l. When the AP device 800 is operating as a device for decompressing IQ data, the PROG 830 is assumed to include program instructions that, when executed by the DP 810, enable the device 800 to operate in accordance with the exemplary embodiments of this disclosure, as discussed herein with the flow diagram as shown in Fig.5.

The MEM 820 may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory, as non-limiting examples. While only one MEM is shown in the device 800, there may be several physically distinct memory units in the device 800. MEM 820 may also be configured to store a codebook, which defines a mapping relationship between set indices and set index codes .

The DP 810 performs any required computations or decisions, which may be involved in the communication procedures as described with reference to Fig.2, Fig. 5. The DP 810 may be of any type suitable to the local technical environment, and may include one or more of general purpose computers, special purpose computers, microprocessors, DSPs and processors based on multi-core processor architecture, as non-limiting examples .

Embodiment 1: a method for IQ, in-phase and quadrature, data compression comprises: receiving an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_d bits ; generating a compressed IQ data block from the uncompressed IQ data block by: generating a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples, and generating a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples; wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and wherein union of a plurality of bit sets constitutes a universal set of N^-bit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

Embodiment 2: according to Embodiment 1, the length of a set index of a specific bit set depends upon element number of the specific bit set. Embodiment 3: according to any of Embodiments 1-2, the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

Embodiment 4: according to any one of Embodiments 1-3, the one or more set index codes are obtained by encoding one or more set indices of the one or more uncompressed samples according to a codebook, wherein the codebook defines a mapping relationship between set indices and set index codes .

Embodiment 5 : according to Embodiment 1 or 2 , generating a compressed IQ data block further comprises: in response to determining that a sum of length of the sign bit segment and length of the set index segment is less than a predetermined length for the compressed IQ data block, generating a residual bit segment of the compressed IQ data by using one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples, such that a sum of the length of the sign bit segment, the length of the set index segment and the length of the residual bit segment is equal to the predetermined length for the compressed IQ data block.

Embodiment 6: according to Embodiment 5, generating a residual bit segment of the compressed IQ data comprises: generating a residual bit segment of the compressed IQ data by using bits of bit sequences from the uncompressed samples in a predefined sequence. Embodiment 7: a method for IQ, in-phase and quadrature, data decompression comprises: receiving a compressed IQ data block including a sign bit segment and a set index segment, the compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; recovering the uncompressed IQ data block from the compressed IQ data block by: recovering one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block, and recovering one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, wherein the set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples; and wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of _V_d-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

Embodiment 8: according to Embodiment 7, wherein length of a set index of a specific bit set depends upon element number of the specific bit set.

Embodiment 9: according to any of Embodiments 7-8, the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

Embodiment 10: according to any of Embodiments 7-9, wherein the one or more set indices of the one or more uncompressed samples are obtained by decoding the one or more set index codes according to a codebook, wherein the codebook defines a mapping relationship between set indices and set index codes.

Embodiment 11: according to any of Embodiments 7-9, the compressed IQ data block further includes a residual segment, and recovering the uncompressed IQ data block further comprises : recovering (S523) one or more second portions of corresponding one or more sequences of the one or more uncompressed samples from the residual bit segment of the compressed IQ data.

Embodiment 12: according to Embodiment 11, recovering an uncompressed IQ data block further comprises: filling a recovered sample with bit "0" or bit "1", in response to determining that an overall length of the first and second portions of the bit sequence of the recovered sample is less than N_d.

Embodiment 13: a compressor for IQ, in-phase and quadrature, data compression, comprises: a receiving unit configured to receive an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; a generating unit configured to generating a compressed IQ data block from the uncompressed IQ data block, the generating unit comprising: a sign bit segment generating unit configured to generate a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples, and a set index segment generating unit configured to generate a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples; wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set..

Embodiment 14 : according to Embodiment 13 , the length of a set index of a specific bit set depends upon element number of the specific bit set.

Embodiment 15: according to Embodiment 13 or 14, the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

Embodiment 16: according to any one of Embodiments 13-15, the one or more set index codes are obtained by encoding one or more set indices of the one or more uncompressed samples according to a codebook, wherein the codebook defines a mapping relationship between set indices and set index codes.

Embodiment 17: according to Embodiment 13 or 14, the generating unit further comprises: a residual bit segment generating unit configured to generate, in response to determining that the sum of the length of the sign bit segment and the length of the set index segment is less than a predetermined length for the compressed IQ data block, a residual bit segment of the compressed IQ data by using one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples, such that a sum of the length of the sign bit segment, the length of the set index segment and the length of the residual bit segment is equal to the predetermined length for the compressed IQ data block.

Embodiment 18: according to Embodiment 17, the residual bit segment generating unit is configured to generate a residual bit segment of the compressed IQ data by using bits of bit sequences from the uncompressed samples in a predefined sequence .

Embodiment 19: a decompressor for IQ, in-phase and quadrature, data decompression, comprises: a receiving unit configured to receive a compressed IQ data block including a sign bit segment and a set index segment , the compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; a recovering unit configured to recover the uncompressed IQ data block from the compressed IQ data block the recovering unit comprising: a sign bit recovering unit configured to recover one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block; a first bit sequence recovering unit configured to recover one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, wherein the set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples, wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

Embodiment 20: according to Embodiment 19, the length of a set index of a specific bit set depends upon element number of the specific bit set.

Embodiment 21: according to Embodiment 19 or 20, the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

Embodiment 22: according to any of Embodiment 19-21, one or more set indices of the one or more uncompressed samples are obtained by decoding the one or more set index codes according to a codebook, wherein the codebook defines a mapping relationship between set indices and set index codes .

Embodiment 23: according to any of Embodiments 19-21, the compressed IQ data block further includes a residual segment, and the recovering unit further comprises: a second bit sequence recovering unit configured to recover one or more second portions of corresponding one or more bit sequences of the one or more uncompressed samples from a residual bit segment of the compressed IQ data.

Embodiment 24 : according to Embodiment 23, the recovering unit is further configured to fill a recovered sample with bit "0" or bit "1", in response to determining that an overall length of the first and second portions of the bit sequence of the recovered sample is less than N_d.

Embodiment 25: an apparatus for IQ, in-phase and quadrature, data compression, comprises one or more processors; and one or more memories including computer program code, the one or more memories and the computer program code configured to, with the one or more processors, cause the apparatus to perform the method according to any of Embodiments 1-6.

Embodiment 26: an apparatus for IQ, in-phase and quadrature, data decompression, comprises one or more processors; and one or more memories including computer program code, the one or more memories and the computer program code configured to, with the one or more processors, cause the apparatus to perform the method according to any of Embodiments 7-12. Embodiment 27: a compressing device for IQ, in-phase and quadrature, data compression, comprises: processing means adapted to: receive an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; generate a compressed IQ data block from the uncompressed IQ data block by: generating a sign bit segment of the compressed IQ data block by using one or more sign bits of the one or more uncompressed samples, and generating a set index segment of the compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples; wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and wherein union of a plurality of bit sets constitutes a universal set of iV_d-bit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set. Embodiment 28 : according to Embodiment 27, the processing means comprise a processor and a memory and wherein the memory is containing instructions executable by the processor.

Embodiment 29: a decompressing device for IQ, in-phase and quadrature, data decompression, comprises: processing means adapted to: receive a compressed IQ data block including a sign bit segment and a set index segment, the compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of the one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; recover the uncompressed IQ data block from the compressed IQ data block by: recovering one or more sign bits of the one or more uncompressed samples from the sign bit segment of the compressed IQ data block, and recovering one or more first portions of corresponding one or more bit sequences of the one or more uncompressed samples based on the set index segment of the compressed IQ data block, wherein the set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in the one or more uncompressed samples; wherein each of the one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying the specific bit set.

Embodiment 30 : according to Embodiment 29, the processing means comprise a processor and a memory and wherein the memory is containing instructions executable by the processor.

In general, the various exemplary embodiments may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. For example, some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the disclosure is not limited thereto. While various aspects of the exemplary embodiments of this disclosure may be illustrated and described as block and signaling diagrams, it is well understood that these blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non- limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof .

As such, it should be appreciated that at least some aspects of the exemplary embodiments of the disclosures may be practiced in various components such as integrated circuit chips and modules. As well known in the art, the design of integrated circuits is by and large a highly automated process . The present disclosure may also be embodied in the computer program product which comprises all features capable of implementing the method as depicted herein and may implement the method when loaded to the computer system. The present disclosure has been specifically illustrated and explained with reference to the preferred embodiments. The skilled in the art should understand various changes thereto in form and details may be made without departing from the spirit and scope of the present disclosure.

Claims

WHAT IS CLAIMED

1. A method (100) for IQ, in-phase and quadrature, data compression, comprising: receiving (S110) an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; generating (S120) a compressed IQ data block from said uncompressed IQ data block by: generating (S121) a sign bit segment of said compressed IQ data block by using one or more sign bits of said one or more uncompressed samples, and generating (S122) a set index segment of said compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples; wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N^; and wherein union of a plurality of bit sets constitutes a universal set of N^-hit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

2. The method (100) according to Claim 1, wherein length of a set index of a specific bit set depends upon element number of said specific bit set.

3. The method (100) according to any of Claims 1-2 , wherein the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

4. The method (100) according to any one of Claims 1-3, wherein said one or more set index codes are obtained by encoding one or more set indices of said one or more uncompressed samples according to a codebook, wherein said codebook defines a mapping relationship between set indices and set index codes .

5. The method (100) according to Claim 1 or 2 , wherein generating a compressed IQ data block further comprises: in response to determining that a sum of length of said sign bit segment and length of said set index segment is less than a predetermined length for said compressed IQ data block, generating (S123) a residual bit segment of said compressed IQ data by using one or more second portions of corresponding one or more bit sequences of said one or more uncompressed samples, such that a sum of the length of said sign bit segment, the length of said set index segment and the length of said residual bit segment is equal to said predetermined length for said compressed IQ data block.

6. The method (100) according to Claim 5, wherein generating a residual bit segment of said compressed IQ data comprises : generating a residual bit segment of said compressed IQ data by using bits of bit sequences from the uncompressed samples in a predefined sequence.

7. A method (500) for IQ, in-phase and quadrature, data decompression, comprising: receiving (S510) a compressed IQ data block including a sign bit segment and a set index segment, said compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_d bits ; recovering (S520) said uncompressed IQ data block from said compressed IQ data block by: recovering (S521) one or more sign bits of said one or more uncompressed samples from said sign bit segment of said compressed IQ data block, and recovering (S522) one or more first portions of corresponding one or more bit sequences of said one or more uncompressed samples based on said set index segment of said compressed IQ data block, wherein said set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples; and wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

8. The method (500) according to Claim 7, wherein length of a set index of a specific bit set depends upon element number of said specific bit set.

9. The method (500) according to any of Claims 7-8, wherein the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

10. The method (500) according to any of Claims 7-9, wherein said one or more set indices of said one or more uncompressed samples are obtained by decoding said one or more set index codes according to a codebook, wherein said codebook defines a mapping relationship between set indices and set index codes .

11. The method (500) according to any of Claims 7-9, wherein said compressed IQ data block further includes a residual bit segment, and recovering said uncompressed IQ data block further comprises : recovering (S523) one or more second portions of corresponding one or more sequences of said one or more uncompressed samples from said residual bit segment of said compressed IQ data.

12. The method (500) according to Claim 11, wherein recovering an uncompressed IQ data block further comprises : filling a recovered sample with bit "0" or bit "1", in response to determining that an overall length of the first and second portions of the bit sequence of said recovered sample is less than Na-

13. A compressor (600) for IQ, in-phase and quadrature, data compression, comprising: a receiving unit (610) configured to receive an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; a generating unit (620) configured to generate a compressed IQ data block from said uncompressed IQ data block, said generating unit comprising: a sign bit segment generating unit (621) configured to generate a sign bit segment of said compressed IQ data block by using one or more sign bits of said one or more uncompressed samples, and a set index segment generating unit (622) configured to generate a set index segment of said compressed IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples; wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N^; and wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

14. The compressor (600) according to Claim 13, wherein length of a set index of a specific bit set depends upon element number of said specific bit set.

15. The compressor (600) according to Claim 13 or 14, wherein the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

16. The compressor (600) according to any one of Claims 13-15, wherein said one or more set index codes are obtained by encoding one or more set indices of said one or more uncompressed samples according to a codebook, wherein said codebook defines a mapping relationship between set indices and set index codes .

17. The compressor (600) according to Claim 13 or 14, wherein said generating unit (620) further comprises: a residual bit segment generating unit ( 623 ) configured to generate, in response to determining that the sum of the length of said sign bit segment and the length of said set index segment is less than a predetermined length for said compressed IQ data block, a residual bit segment of said compressed IQ data by using one or more second portions of corresponding one or more bit sequences of said one or more uncompressed samples, such that a sum of the length of said sign bit segment, the length of said set index segment and the length of said residual bit segment is equal to said predetermined length for said compressed IQ data block.

18. The compressor (600) according to Claim 17, wherein said residual bit segment generating unit ( 623 ) is configured to generate a residual bit segment of said compressed IQ data by using bits of bit sequences from the uncompressed samples in a predefined sequence.

19. A decompressor (700) for IQ, in-phase and quadrature , data decompression, comprising: a receiving unit (710) configured to receive a compressed IQ data block including a sign bit segment and a set index segment, said compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; a recovering unit (720) configured to recover said uncompressed IQ data block from said compressed IQ data block, said recovering unit comprising: a sign bit recovering unit (721) configured to recover one or more sign bits of said one or more uncompressed samples from said sign bit segment of said compressed IQ data block; and a first bit sequence recovering unit (722) configured to recover one or more first portions of corresponding one or more bit sequences of said one or more uncompressed samples based on said set index segment of said compressed IQ data block, wherein said set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples, wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

20. The decompressor (700) according to Claim 19 , wherein length of a set index of a specific bit set depends upon element number of said specific bit set.

21. The decompressor (700) according to Claim 19 or 20, wherein the length of the set index code M₀ has a smaller value, if distribution probability of data in the corresponding bit set is higher.

22. The decompressor (700) according to any of Claims 19-21, wherein one or more set indices of said one or more uncompressed samples are obtained by decoding said one or more set index codes according to a codebook, wherein said codebook defines a mapping relationship between set indices and set index codes .

23. The decompressor (700) according to any of Claims 19-21, wherein said compressed IQ data block further includes a residual bit segment, and said recovering unit (720) further comprises: a second bit sequence recovering unit (723) configured to recover one or more second portions of corresponding one or more bit sequences of said one or more uncompressed samples from a residual bit segment of said compressed IQ data.

24. The decompressor (700) according to Claim 23 , wherein said recovering unit (720) is further configured to fill a recovered sample with bit "0" or bit "1", in response to determining that an overall length of the first and second portions of the bit sequence of said recovered sample is less than N_d.

25. An apparatus for IQ, in-phase and quadrature, data compression, comprising one or more processors; and one or more memories including computer program code, the one or more memories and the computer program code configured to, with the one or more processors, cause said apparatus to perform the method according to any of Claims 1-6.

26. An apparatus for IQ, in-phase and quadrature, data decompression, comprising one or more processors; and one or more memories including computer program code, the one or more memories and the computer program code configured to, with the one or more processors, cause said apparatus to perform the method according to any of Claims 7-12.

27. A compressing device for IQ, in-phase and quadrature, data compression, comprising: processing means adapted to: receive an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; generate a compressed IQ data block from said uncompressed IQ data block by: generating a sign bit segment of said compressed IQ data block by using one or more sign bits of said one or more uncompressed samples, and generating a set index segment of said compressed

IQ data block by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples; wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and wherein union of a plurality of bit sets constitutes a universal set of N_d-bit sequences, and each bit sequence belonging to a specific bit set has its first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

28. The compressing device according to Claim 27, wherein the processing means comprise a processor and a memory and wherein said memory is containing instructions executable by said processor.

29. A decompressing device for IQ, in-phase and quadrature, data decompression, comprising: processing means adapted to: receive a compressed IQ data block including a sign bit segment and a set index segment, said compressed IQ data block corresponding to an uncompressed IQ data block including one or more uncompressed samples, wherein each of said one or more uncompressed samples includes a bit sequence and at least one sign bit, the bit sequence having a length of N_dbits; recover said uncompressed IQ data block from said compressed IQ data block by: recovering one or more sign bits of said one or more uncompressed samples from said sign bit segment of said compressed IQ data block, and recovering one or more first portions of corresponding one or more bit sequences of said one or more uncompressed samples based on said set index segment of said compressed IQ data block, wherein said set index segment is generated by using one or more set index codes corresponding to one or more bit sequences included respectively in said one or more uncompressed samples; wherein each of said one or more set index codes indicates a bit set to which a bit sequence of a respective uncompressed sample belongs, and length of the set index code M₀ is less than length of the corresponding bit sequence N_d; and, wherein union of a plurality of bit sets constitutes a universal set of i¾-bit sequences, and each bit sequence belonging to a specific bit set has it first portion with a bit pattern that serves as a set index uniquely identifying said specific bit set.

30. The decompressing device according to Claim 29, wherein the processing means comprise a processor and a memory and wherein said memory is containing instructions executable by said processor.