WO2013010437A1 - 一种数据压缩方法和设备 - Google Patents
一种数据压缩方法和设备 Download PDFInfo
- Publication number
- WO2013010437A1 WO2013010437A1 PCT/CN2012/078061 CN2012078061W WO2013010437A1 WO 2013010437 A1 WO2013010437 A1 WO 2013010437A1 CN 2012078061 W CN2012078061 W CN 2012078061W WO 2013010437 A1 WO2013010437 A1 WO 2013010437A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- input signal
- quantization
- information
- bit width
- quantized
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/04—Protocols for data compression, e.g. ROHC
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
Definitions
- the present invention relates to the field of communications technologies, and in particular, to a data compression method and apparatus. Background technique
- a base station device (such as an eNB) is a distributed base station device, and a base station device is a BBU (Building Base Band Unit).
- the processing unit) and the RRU (Radio Remote Unit) are a combination of base stations that can be flexibly and distributedly installed.
- FIG. 1 it is a schematic diagram of a base station device composed of an RRU and a BBU, and the RRU and the BBU.
- the interface between the two is an Ir interface, that is, the RRU is connected to the BBU through the Ir interface, and transmits data between the two.
- the Ir interface uses a transmission medium such as an optical fiber. If the data of the Ir interface is compressed, the demand for the transmission medium can be greatly reduced.
- the data compression mode of the Ir interface in the current LTE system is: by performing automatic gain adjustment on the input signal, controlling the dynamic range of the signal, correspondingly reducing the quantization bit width of the signal, and quantifying the quantization algorithm.
- This method can compress 16-bit wide data to 12-bit based on the reliability of the signal, and the ratio before and after compression is 4:3.
- the ratio of the size of the data before and after compression is the compression ratio.
- the embodiment of the invention provides a data compression method and device, which compresses data reasonably and optimizes compression performance.
- an embodiment of the present invention provides a data compression method, including:
- the transmitting end quantizes the input signal according to the adaptive quantization information
- the transmitting end compresses the quantized input signal
- the transmitting end sends the compressed input signal to the receiving end.
- the embodiment of the invention provides a data compression device, including:
- a quantization module configured to quantize the input signal according to the adaptive quantization information
- a compression module configured to compress the quantized input signal
- the sending module is configured to send the compressed input signal to the receiving end.
- the present invention has at least the following advantages:
- FIG. 1 is a schematic diagram of a base station device composed of a RRU and a BBU in the prior art
- FIG. 2 is a schematic structural diagram of a data transmission device according to Embodiment 1 of the present invention
- FIG. 3 is a flowchart of a data compression method according to Embodiment 1 of the present invention
- 4 is a block diagram of a segmentation compression implementation in the first embodiment of the present invention
- FIG. 5 is a schematic structural diagram of a data transmission device according to Embodiment 2 of the present invention.
- FIG. 6 is a schematic structural diagram of a data compression device according to Embodiment 3 of the present invention.
- the LTE/LTE-A system adopts a technology such as a large bandwidth and multiple antennas, which results in a large amount of data transmitted on the Ir interface, and has a low compression ratio.
- the embodiment of the present invention provides a data compression method. And the device, through adaptive quantization and segmentation compression of the signal, thereby realizing compression of the transmission data, optimizing signal transmission, and improving the efficiency of system data transmission.
- Embodiment 1 of the present invention provides a data (that is, a signal, including but not limited to a signal of a 3G/4G system) compression method, which is applied to a data transmission system including a transmitting end and a receiving end.
- a data transmission system including a transmitting end and a receiving end.
- the transmitting end can also be the receiving end, so each data transmission device can have the structure shown in FIG. 2.
- the power/amplitude adjuster, the adaptive quantizer, the segmentation compressor and the buffer controller are functional modules of the data transmission device as a transmitting end; the power/amplitude compensator, the dequantizer, the de segmentation compressor and the demodulator are The data transmission device acts as a functional module of the receiving end.
- Each of the above functional modules can be used alone or in combination.
- the method is applicable to a base station device including a BBU and an RRU, and is used to compress data transmitted by an Ir interface between a BBU and an RRU.
- BBU When the data is sent to the RRU, the BBU is the sender and the RRU is the receiver.
- the RRU sends data to the BBU, the BBU is the receiver and the RRU is the sender.
- the data compression method includes the following steps:
- Step 301 The transmitting end adjusts the power or amplitude of the input signal (the input signal is input data).
- the transmitter can perform AGC (Automatic Gain Control) adjustment on the input signal.
- AGC Automatic Gain Control
- the input signal may be first AGC adjusted. Since the AGC target value has a great influence on the system performance after quantization, different AGC target values will affect the signal distribution characteristics and affect the compression performance of the signal. Therefore, it is necessary to select the AGC target value reasonably to obtain the optimal system performance. For example, when the AGC target value is large, the probability density of the large signal will be large, so a smaller quantization interval can be selected in the large signal region, and conversely, a smaller quantization interval is selected in the small signal region.
- the function module performing this step may be the power/amplitude compensator in FIG. 2.
- Step 302 The transmitting end quantizes the input signal according to the adaptive quantization information.
- the transmitting end may quantize the input signal after power or amplitude adjustment according to the adaptive quantization information.
- the adaptive quantization information includes but is not limited to one or any combination of the following: a quantization mode (ie, a quantization algorithm); a quantization parameter; a quantization bit width (ie, a quantized bit width).
- a quantization mode ie, a quantization algorithm
- a quantization parameter ie, a quantization parameter
- a quantization bit width ie, a quantized bit width
- the adaptive quantization information is a quantization method.
- the transmitting end determines the quantization mode according to the distribution characteristics of the input signal, and quantizes the input signal by the quantization method.
- the transmitting end determines that the quantization mode is a uniform quantizer, that is, the equalization quantizer is used for quantization.
- the transmitting end determines that the quantization mode is non-uniform Quantizer, that is, when the probability density characteristic of the input signal is non-homogeneously distributed, the transmitting end determines a non-uniform hook quantizer that can match the probability density function of the source, such as the non-uniformity that can be used for the speech signal.
- the hook quantizer is for A law or law.
- the adaptive quantization information is the quantization bit width.
- the transmitting end monitors the buffer status and adjusts the quantized bit width according to the buffer status. Wherein, when the buffer state is lower than the preset first threshold, the transmitting end increases the quantized bit width; when the buffer state is higher than the preset second threshold, the transmitting end decreases the quantized bit width.
- the preset first threshold is smaller than the preset second threshold, and the preset first threshold and the preset second threshold may be selected according to actual experience values.
- the buffer control before the data (i.e., the corresponding signal) is sent to the receiving end, in order to ensure the constant and continuous signal transmission rate, the buffer control can also be performed.
- the function module that performs the related operations can be the cache controller in Figure 2.
- the buffer controller ensures a constant signal transmission rate through the output port of the control signal.
- the adaptive quantizer or the segmentation compressor can be notified by feedback, and the corresponding processing method is changed by the adaptive quantizer or the segmentation compressor.
- the adaptive quantizer changes the quantization mode, the quantization bit width or the quantization parameter
- the segmentation compressor changes the segmentation mode.
- the partial signal performance is sacrificed.
- the increase of the compression ratio in this case, even if the quantization error is introduced, the overall signal loss is small, and only the individual signals are sacrificed, and the quantization error of most of the signals is small and the performance is good.
- the sender can monitor the status of the cache in real time.
- the quantized bit width can be adjusted according to the state of the buffer, when the buffer is empty (When the buffer state is lower than the preset first threshold), the quantized bit width can be increased, the compression ratio can be reduced, thereby reducing the quantization error; when the buffer is full (the buffer state is higher than the preset second threshold), it can be lowered The quantized bit width is increased, the compression ratio is increased, and the accuracy of the partial signal is sacrificed so that the signal does not overflow.
- the adaptive quantization information is a quantization parameter.
- the adjustment of the quantization parameters will also affect the compression ratio of the signal, which in turn affects the buffer state.
- different codebook changes will affect the compression ratio of the segmentation compression.
- the quantization parameter may be adjusted according to the state of the buffer (illustrated by using codebook quantization as an example).
- codebook quantization Illustrated by using codebook quantization as an example.
- a preferred codebook may be selected for quantization to reduce the compression ratio. , thereby reducing the quantization error; when the buffer is full, the poor codebook can be selected for quantization to increase the compression ratio, and the accuracy of the partial signal is sacrificed so that the signal does not overflow.
- the function module for performing this step may be the adaptive quantizer in FIG. 2 .
- Step 303 The transmitting end compresses the quantized input signal.
- the transmitting end may determine the data segment information, and segment the quantized input signal by using the data segment information, and compress the quantized input signal by using the same bit width information in each segment.
- the data can be encoded with a suitable bit width; for example, each data is treated as a single data segment. Since the receiving end needs to know the bit width information to complete the decompression of the data, it is necessary to simultaneously transmit the bit width information; and since the bit width information occupies a large amount of resources, the size of the data segment can be appropriately increased to reduce the bit width information. The proportion of resources. Further, when the size of the data segment is increased, there is no guarantee that each data is represented by the most suitable bit width, that is, some small data wastes resources when expressed by the bit width bits of the segment, and also introduces a year-on-year increase. Processing delay.
- the size of the data segment can be adaptively selected by simulating and analyzing the signal, and the size of the segment will directly affect the compression ratio.
- the adaptive quantization information can be transmitted through multiplexing with data.
- the data and the bit width information may be processed by using a certain encoding manner, that is, the transmitting end may encode the bit width information and the quantized input signal by using a specified encoding manner. For example, specify the encoding method as differential encoding.
- the block compression implementation block diagram the specific implementation steps include:
- Step (1) dividing the codebook subscript data quantized by each radio frame into blocks, and the number of samples included in each block is;
- Step (2) differentially encoding data of each data block
- Step (3) segment each data block, and the number of samples included in each segment is;
- Step (4) compress and encode each data segment, the specific process is: searching for the maximum value of the absolute value in the data segment, Representing and finding the corresponding quantization bit width, that is, where encoding is performed as bit width information.
- the number of bits occupied by the bit width information can be reduced according to the characteristics of the signal, and the value of the signal bit width is mostly n, n-1, n-2, n. -3, only when the individual value is n-4, the bit width of the data segment with the bit width n-4 can be limited to n-3, so that the bit width information can be represented by 2 bits (n is greater than 4) Positive integer).
- the optional bit width information may be transmitted by absolute coding or differential coding; if it is required to improve the accuracy, a certain redundancy coding protection mechanism (such as repeated coding, etc.) may be introduced, and the bit width information may be added. It is multiplexed with the data and transmitted. Further, if there is adaptive quantization information and variable segmentation information, it is also necessary to multiplex with the bit width information and the data portion.
- bit width information and adaptive information and variable segmentation information can be placed in the cyclic prefix of the signal, thereby reducing the impact on the data.
- the above-mentioned methods of block differential coding and segmentation compression are all processes of lossless compression. If the compression ratio is low, the method cannot meet the requirements. It is also possible to introduce a certain lossy processing on the data before the compression processing, for example, introducing an attenuation factor to reduce the width of the effective sample, etc.; further, the data width of the segmented compression is a variable value, so the scheme is a non- A compression scheme with a constant compression ratio.
- function module for performing this step may be the segment compressor in FIG. 2.
- Step 304 The sender buffers the compressed input signal.
- the buffer control may be performed, that is, the output port of the control signal is used to ensure the signal transmission rate is constant. Further, if the compression ratio of the current frame signal is lower than the compression ratio controlled by the output port, the retained signal needs to be buffered for storage, and the port is transmitted to continue transmitting the data. In addition, it is also necessary to monitor the storage state of the cache in real time. When there are more residual signals in the buffer, the adaptive quantizer or the segmentation compressor can be notified by feedback, and the corresponding processing mode is changed by the adaptive quantizer or the segmentation compressor.
- function module that performs this step may be the cache controller in FIG. 2.
- Step 305 The sending end sends the compressed input signal to the receiving end.
- the sender transmits the compressed input signal to the receiving end at a signal rate that satisfies a preset policy (such as a policy in which the data rate remains constant) for the buffered compressed input signal.
- a preset policy such as a policy in which the data rate remains constant
- Step 306 The receiving end receives the compressed input signal, and performs decoupling, decompression, de-quantization, and power compensation on the received data.
- the foregoing process is an inverse process of the transmitting end, and details are not described herein again.
- the scheme can support a larger dynamic range of the input signal, and the adjusted signal is more suitable for quantization.
- adaptive quantization is achieved by changing the quantization algorithm, quantization parameters and quantization bit width, and the quantization algorithm is more efficient and more designed. Reasonable.
- the segmentation compression scheme may be optionally used to further increase the compression ratio.
- the relevant information is fed back to the adaptive quantizer or the segmentation compressor, and the compression ratio is adjusted by adjusting the quantization information and the segmentation information.
- the non-constant compression ratio data is processed by the cache controller to output at a constant rate, and the state of the buffer is adjusted in real time and the compression ratio is adjusted to achieve a constant rate output of the data.
- the receiving end adopts corresponding demodulation processing.
- a second embodiment of the present invention provides a data compression method, which is described by taking an application in an Ir interface data compression scheme of an LTE system as an example, as shown in FIG.
- the power/amplitude adjuster is AGC and the power/amplitude compensator is AGC compensation.
- the Ir interface protocol supports transmission of all valid data between the BBU and the RRU.
- the maximum IQ data sampling width is 16 bits, and the optical fiber transmission needs 8B/10B encoding, that is, the link encoding mode of the encoding of the underlying transmission device serdes.
- the purpose is to make the link transmit 0 and 1 numbers equalized, and the receiving end can recover the clock from the link.
- the technical solutions provided by the embodiments of the present invention may be used for the LTE system signal, and the specific implementation includes:
- Step (1) the signal is subjected to AGC processing to adjust the amplitude of the signal.
- Step (2) storing the codebook, the data bit width in the codebook is 16 bits, which is consistent with the data bit width after the AGC.
- Step (3) quantizing the signal using the generated quantized codebook.
- the quantized codebook is trained according to the distribution characteristics of the signal.
- Step (4) compressing the codebook subscript by means of differential encoding and segmentation compression, and transmitting the fixed Mbit via the optical fiber through the buffer control.
- a feedback mechanism is introduced, and the compression ratio is calculated in real time by monitoring the size of the remaining space of the buffer, and fed back to the adaptive quantizer. If the current compression ratio is low, there are several ways to improve it:
- the segmentation information is encoded, a table of segmentation information is established, and the table interval subscript and bit width information and data corresponding to the segmentation information are multiplexed and transmitted. .
- the bit width information, the adaptive quantization information, and the segmentation information may be placed in a cyclic prefix of the signal and multiplexed with the data for transmission.
- Step (5) At the receiving end, the bit data is first subjected to segmentation decompression, and the difference is solved, and then the codebook is used to look up the corresponding data by means of table lookup, and decompression, decoding, and AGC factor recovery are completed.
- Embodiment 3 is effective to compress 16-bit wide data to Mbit (0 ⁇ M ⁇ 16).
- the embodiment of the present invention further provides a data compression device.
- the device includes:
- the quantization module 11 is configured to quantize the input signal according to the adaptive quantization information
- the compression module 12 is configured to compress the quantized input signal
- the sending module 13 is configured to send the compressed input signal to the receiving end.
- the adaptive quantization information includes one or any combination of the following: a quantization mode; a quantization parameter; a quantization bit width.
- the quantization module 11 is specifically configured to: when the adaptive quantization information is in a quantization mode, determine a quantization mode according to a distribution characteristic of the input signal, and quantize the input signal by using the quantization mode.
- the quantization module 11 is further configured to: when the input signal is a uniform hook, determine that the quantization mode is a uniform quantizer; and when the input signal is non-uniformly distributed, determine that the quantization mode is non-uniform quantization Device.
- the quantization module 11 is specifically configured to monitor a buffer state when the adaptive quantization information is a quantization bit width, and adjust the quantized bit width according to the buffer state.
- the quantization module 11 is further configured to: when the buffer state is lower than a preset first threshold, increase the quantized bit width; when the buffer state is higher than a preset second threshold, decrease the The quantized bit width.
- the quantization module 11 is specifically configured to perform power or amplitude adjustment on the input signal, and quantize the adjusted input signal according to the adaptive quantization information.
- the compression module 12 is specifically configured to determine data segment information, and segment the quantized input signal by using the data segment information, and use the same bit width information for each of the quantized inputs in each segment The signal is compressed.
- the compression module 12 is further configured to perform encoding processing on the bit width information and the quantized input signal by using a specified encoding manner.
- the sending module 13 is specifically configured to cache the compressed input signal, and send the compressed input signal to the signal rate of the preset policy. Said receiver.
- modules of the device of the present invention may be integrated or may be deployed separately.
- the above modules can be combined into one module, or they can be further split into multiple sub-modules.
- the present invention can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is a better implementation. the way.
- the technical solution of the present invention which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for making a A computer device (which may be a personal computer, server, or network device, etc.) performs the methods described in various embodiments of the present invention.
- modules in the apparatus in the embodiment may be distributed in the apparatus of the embodiment according to the description of the embodiment, or may be correspondingly changed in one or more apparatuses different from the embodiment.
- the modules of the above embodiments may be combined into one module, or may be further split into multiple sub-modules.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
本发明公开了一种数据压缩方法和设备,该方法包括:发送端根据自适应量化信息对输入信号进行量化;所述发送端对量化后的输入信号进行压缩;所述发送端将压缩后的输入信号发送给接收端。本发明实施例中,通过对数据进行自适应量化及分段压缩,从而可实现传输数据的压缩,优化数据的传输,并提高系统数据传输的效率。
Description
一种数据压缩方法和设备 本申请要求于 2011 年 7 月 18 日提交中国专利局, 申请号为 201110200699.2, 发明名称为 "一种数据压缩方法和设备"的中国专 利申请的优先权, 其全部内容通过引用结合在本申请中。 技术领域
本发明涉及通信技术领域, 尤其涉及一种数据压缩方法和设 备。 背景技术
在 LTE ( Long Term Evolution , 长期演进 ) /LTE-A ( LTE- Advanced, 高级 LTE ) 系统中, 基站设备(如 eNB ) 为分 布式基站设备, 且基站设备是由 BBU ( Building Base band Unit, 基带处理单元)、 RRU ( Radio Remote Unit, 射频远端单元 )构成, 是一种可以灵活分布式安装的基站组合, 如图 1所示, 为 RRU和 BBU组成的基站设备的示意图, RRU与 BBU之间的接口为 Ir接 口, 即 RRU通过 Ir接口与 BBU相连, 并传输二者之间的数据。
现有技术中, Ir接口采用光纤等传输媒质, 如果对 Ir接口的 数据进行压缩, 则可以大大减少传输媒质的需求量。 当前 LTE系 统中 Ir接口的数据压缩方式为: 通过对输入信号进行自动增益调 节, 控制信号的动态范围, 相应的降低信号的量化位宽、 量化算 法为均勾量化。 该方式可以在保证信号可靠性的基础上, 将 16bit 位宽的数据压缩至 12bit, 压缩前后的比率为 4: 3。 其中, 压缩前 后的数据大小之比为压缩比。
在实现本发明的过程中, 发明人发现现有技术中至少存在以 下问题:
现有技术的压缩比低, 且算法的通用性不强, 当输入信号为 均匀分布时, 均匀量化是最佳的量化器; 但当输入信号为非均匀
分布式时, 均勾量化对量化电平的分配不够合理, 无法充分去除 信号中的冗余成分。 发明内容
本发明实施例提供一种数据压缩方法和设备, 以合理的对数 据进行压缩, 优化压缩性能。
为了达到上述目的, 本发明实施例提供一种数据压缩方法, 包括:
发送端根据自适应量化信息对输入信号进行量化;
所述发送端对量化后的输入信号进行压缩;
所述发送端将压缩后的输入信号发送给接收端。
本发明实施例提供一种数据压缩设备, 包括:
量化模块, 用于根据自适应量化信息对输入信号进行量化; 压缩模块, 用于对量化后的输入信号进行压缩;
发送模块, 用于将压缩后的输入信号发送给接收端。
与现有技术相比, 本发明至少具有以下优点:
通过对数据进行自适应量化及分段压缩, 从而可实现传输数 据的压缩, 优化数据的传输, 并提高系统数据传输的效率。 附图说明
图 1是现有技术中 RRU和 BBU组成的基站设备的示意图; 图 2是本发明实施例一中的数据传输设备的结构示意图; 图 3是本发明实施例一提供的一种数据压缩方法流程示意图; 图 4是本发明实施例一中分段压缩实现框图;
图 5是本发明实施例二中的数据传输设备的结构示意图; 图 6是本发明实施例三提供的一种数据压缩设备结构示意图。 具体实施方式
发明人在实现本发明的过程中注意到: 度量通信的技术性能
是从通信的数量和质量来讨论的, 数量指标用有效性度量, 质量 指标用可靠性度量。 在保证通信质量的同时, 通过寻找提高系统 有效性的方式, 来保证在有限的信道容量下能够传输更多的信号。 其中, 信号若不经过处理, 则会存在大量的冗余成分, 因此可通 过量化和压缩等方式去掉多余成分, 实现有效性的提升。
实际应用中, 由于量化会产生量化误差, 并引入失真, 因此 需要在限定失真条件下寻找所需比特最少的量化压缩编码方式。
针对现有技术中, LTE/LTE-A 系统采用大带宽、 多天线等技 术, 从而导致 Ir接口传输的数据量较大, 现有压缩比低等问题, 本发明实施例提供一种数据压缩方法和设备, 通过对信号进行自 适应量化及分段压缩, 从而实现传输数据的压缩, 优化信号传输, 提高系统数据传输的效率。
下面将结合本发明中的附图, 对本发明中的技术方案进行清 楚、 完整地描述, 显然, 所描述的实施例仅仅是本发明的一部分 实施例, 而不是全部的实施例。 基于本发明中的实施例, 本领域 普通技术人员在没有做出创造性劳动前提下所获得的所有其他实 施例, 都属于本发明保护的范围。
实施例一
本发明实施例一提供一种数据(即信号, 包括但不限于 3G/4G 系统的信号) 压缩方法, 应用于包括发送端和接收端的数据传输 系统中, 对于每个数据传输设备, 其可以为发送端, 也可以为接 收端, 因此各数据传输设备可以如图 2所示的结构。
其中, 功率 /幅度调整器、 自适应量化器、 分段压缩器和緩存 控制器是数据传输设备作为发送端的功能模块;功率 /幅度补偿器、 解量化器、 解分段压缩器和解控器是数据传输设备作为接收端的 功能模块。 上述各功能模块可单独使用, 也可选择性使用其中几 个。
例如, 该方法可应用于包括 BBU和 RRU的基站设备中, 并 用于对 BBU和 RRU之间的 Ir接口所传输的数据进行压缩。当 BBU
向 RRU发送数据时, 则 BBU为发送端, RRU为接收端; 当 RRU 向 BBU发送数据时, 则 BBU为接收端, RRU为发送端。
基于图 2所示的结构, 如图 3所示, 该数据压缩方法包括以 下步骤:
步骤 301 , 发送端对输入信号(该输入信号为输入数据 )进行 功率或幅度的调整。 例如, 发送端可以对输入信号进行 AGC ( Automatic Gain Control, 自动增益控制)调整。
为了使得本发明实施例可支持更大的输入信号动态范围, 并 使得调整后的信号更加适于量化, 可以先对输入信号进行 AGC调 整。 由于 AGC 目标值对量化后系统性能影响很大, 不同的 AGC 目标值会影响信号的分布特性, 从而影响信号的压缩性能, 因此 需要合理的选择 AGC 目标值以获得最优的系统性能。 例如, 当 AGC 目标值较大时, 大信号的概率密度将会较大, 因此可在大信 号区域选择较小的量化间隔, 反之, 在小信号区域选择较小的量 化间隔。
需要说明的是, 执行本步骤的功能模块可以为图 2中的功率 / 幅度补偿器。
步骤 302, 发送端根据自适应量化信息对输入信号进行量化。 其中, 发送端可根据自适应量化信息对经过功率或幅度调整后的 输入信号进行量化。
本发明实施例中, 该自适应量化信息包括但不限于以下之一 或任意组合: 量化方式 (即量化算法); 量化参数; 量化位宽 (即 量化后的比特位宽)。
( 1 ) 自适应量化信息为量化方式。
当自适应量化信息为量化方式时, 发送端根据输入信号的分 布特性确定量化方式, 并通过量化方式对输入信号进行量化。
其中, 当输入信号为均勾分布时, 发送端确定量化方式为均 匀量化器, 即采用均勾量化器进行量化。
当输入信号为非均匀分布时, 发送端确定量化方式为非均匀
量化器, 即当输入信号的概率密度特性为非均勾分布时, 发送端 确定可采用量化特性与信源的概率密度函数相匹配的非均勾量化 器, 如对于语音信号可采用的非均勾量化器了为 A 律或 律等。
( 2 ) 自适应量化信息为量化位宽。
当自适应量化信息为量化位宽时, 发送端监控緩存状态, 并 根据緩存状态调整量化后的比特位宽。 其中, 当緩存状态低于预 设第一阈值时, 发送端提高量化后的比特位宽; 当緩存状态高于 预设第二阈值时, 发送端降低量化后的比特位宽。 该预设第一阈 值小于预设第二阈值, 且预设第一阈值和预设第二阈值均可以根 据实际经验值进行选择。
本发明实施例中, 在将数据 (即相应的信号) 发送给接收端 之前, 为了保证信号传输速率的恒定和连续性, 还可以进行緩存 控制。 执行相关操作的功能模块可以为图 2 中的緩存控制器。 其 中, 该緩存控制器通过控制信号的输出端口, 来保证信号传输速 率的恒定。
进一步的, 如果当前帧信号的压缩比低于输出端口控制的压 缩比时, 需要将滞留的信号采用緩存进行保存, 并等端口传完信 号, 以继续传输这些数据。 此外, 还需要实时监控緩存的存储状 态, 当緩存中滞留信号较多时, 可通过反馈通知自适应量化器或 分段压缩器, 由自适应量化器或分段压缩器改变相应的处理方式
(如自适应量化器改变量化方式、 量化位宽或量化参数, 分段压 缩器改变分段方式), 后续过程中通过优化量化分段算法或引入一 定量化误差, 牺牲部分信号性能的方式, 实现压缩比的提升; 这 种情况下, 即使引入量化误差, 也使得整体信号损失较小, 只是 牺牲个别信号, 大部分信号的量化误差都很小, 性能较好。
综上所述, 针对该緩存机制, 发送端可实时监控緩存的状态
(如可由緩存控制器实时监控緩存的状态, 并反馈给自适应量化 器), 并且可根据緩存的状态调整量化后的比特位宽, 当緩存较空
(緩存状态低于预设第一阈值) 时, 可以提高量化后的比特位宽, 降低压缩比, 从而减少量化误差; 当緩存较满 (緩存状态高于预 设第二阈值) 时, 可以降低量化后的比特位宽, 提高压缩比, 牺 牲部分信号的精确度使得信号不会溢出。
( 3 ) 自适应量化信息为量化参数。
量化参数的调整也将影响信号的压缩比, 进而影响到緩存状 态, 例如, 对于码本量化, 不同码本的改变将会影响分段压缩的 压缩比。 本发明实施例中, 针对上述緩存机制, 可根据緩存的状 态调整量化参数(以码本量化为例进行说明), 当緩存较空时, 可 以选择较优的码本进行量化, 以降低压缩比, 从而减少量化误差; 当緩存较满时, 可以选择较差的码本进行量化, 以提高压缩比, 牺牲部分信号的精确度使得信号不会溢出。
需要说明的是, 执行本步骤的功能模块可以为图 2 中的自适 应量化器。
步骤 303 , 发送端对量化后的输入信号进行压缩。 其中, 发送 端可确定数据段信息, 并利用数据段信息对量化后的输入信号进 行分段, 在每段内采用相同的位宽信息对量化后的输入信号进行 压缩。
为了使得每个数据均能够使用合适的位宽表示, 可以对该数 据进行合适位宽的编码; 例如, 将每个数据单独作为一个数据段。 由于接收端需要获知位宽信息才能完成数据的解压缩, 因此需要 同时传输位宽信息; 而且由于位宽信息占用大量的资源, 因此可 适当的增加数据段的大小来减小位宽信息所占的资源比例。 进一 步的, 增加数据段的大小时无法保证每个数据均使用最合适的位 宽来表示, 即某些小的数据在用该段的位宽比特表示时会浪费一 些资源, 同时也会引入同比增长的处理时延。
综上所述, 数据段的大小可以通过仿真和分析信号进行自适 应选取, 分段的大小将会直接影响压缩比。
本发明实施例中, 自适应量化信息可通过和数据的复用传输
至接收端。 因此, 为了进一步的提高压缩比, 可以采用一定的编 码方式对数据及位宽信息进行处理, 即发送端可利用指定编码方 式对位宽信息以及量化后的输入信号进行编码处理。 例如, 指定 编码方式为差分编码。
进一步的, 为了降低差分编码的错误传播, 还可采用先分数 据块差分编码, 再分段数据压缩的方式。 如图 4 所示的分段压缩 实现框图, 具体的实现步骤包括:
步骤 ( 1 )、 将每个无线帧量化后的码本下标数据进行分块, 每块包含的样点数为 ;
步骤 (2 )、 将每个数据块的数据进行差分编码;
步骤 (3 )、 将每个数据块进行分段, 每段包含的样点数为 ; 步骤 (4 )对每个数据段进行压缩编码, 具体过程为: 搜索出数据段中绝对值最大值,通过 表示,并求出对应于 的 量化位宽 , 即 ; 其中, 进行编码即为位宽信息。
本发明实施例中, 可根据信号的特点, 通过有效的位宽限制, 降低位宽信息所占的比特数,例如,信号位宽的取值大多为 n, n-1 , n-2, n-3 , 只有个别取值为 n-4时, 则可以通过将位宽为 n-4的数 据段的位宽限制为 n-3 , 使得位宽信息用 2比特表示即可 (n为大 于 4的正整数)。
此外, 可以根据信号的特点, 可选的对位宽信息采用绝对编 码或差分编码传输; 如果需要提高准确度, 也可以引入一定的冗 余编码保护机制 (如重复编码等), 将位宽信息和数据部分复用, 进行传输。 进一步的, 如果存在自适应量化信息和可变分段信息, 则也需要与位宽信息和数据部分进行复用。
其中, 复用的方式有多种, 如对于 LTE信号, 可以将位宽信 息和自适应信息及可变分段信息放置于信号的循环前缀中, 从而 降低对数据的影响。
需要说明的是, 上述的分块差分编码和分段压缩的方案均为 无损压缩的过程, 如果在压缩比很低, 该方式无法满足要求时,
还可以在压缩处理之前对数据引入一定的有损处理, 例如引入衰 减因子减小有效样本的宽度等; 此外, 分段压缩后的数据位宽为 一可变值, 因此该方案是一种非恒定压缩比的压缩方案。
需要说明的是, 执行本步骤的功能模块可以为图 2 中的分段 压缩器。
步骤 304, 发送端对压缩后的输入信号进行緩存。
本发明实施例中, 在将数据发送给接收端之前, 为了保证信 号传输速率的恒定和连续性, 还可以进行緩存控制, 即通过控制 信号的输出端口, 来保证信号传输速率的恒定。 进一步的, 如果 当前帧信号的压缩比低于输出端口控制的压缩比时, 需要将滞留 的信号采用緩存进行保存, 并等端口传完信号, 以继续传输这些 数据。 此外, 还需要实时监控緩存的存储状态, 当緩存中滞留信 号较多时, 可通过反馈通知自适应量化器或分段压缩器, 由自适 应量化器或分段压缩器改变相应的处理方式。
需要说明的是, 执行本步骤的功能模块可以为图 2 中的緩存 控制器。
步骤 305 , 发送端将压缩后的输入信号发送给接收端。 其中, 针对緩存后的压缩后的输入信号, 发送端将以满足预设策略 (如 数据速率保持恒定的策略) 的信号速率将压缩后的输入信号发送 给接收端。
步骤 306, 接收端接收压缩后的输入信号, 并对接收的数据进 行解控, 解压缩和解量化及功率补偿, 上述过程分别为发送端的 逆过程, 在此不再赘述。
综上所述, 本发明实施例中:
( 1 )通过对输入信号进行功率或幅度的调整, 使得该方案可 支持更大的输入信号动态范围, 也使得调整后的信号更加适于量 化。
( 2 )通过使用自适应量化方案, 通过改变量化算法, 量化参 数和量化位宽来实现自适应量化, 量化算法效率更好, 设计更加
合理。
( 3 )通过使用分段压缩方案, 通过对数据的分段, 使得尽量 多的数据均用其合适的位宽表示, 减少信号的冗余。 并且可选择 性的对数据、 位宽信息采用一定编码的方式 (如差分编码), 进一 步的提高压缩比。
( 4 )通过自适应量化方案和分段压缩方案的结合, 在自适应 量化方案执行结束后, 可选择的使用分段压缩方案, 以进一步的 提高压缩比。
( 5 )通过实时的检测当前的压缩比, 将相关信息反馈给自适 应量化器或分段压缩器, 并通过调节量化信息及分段信息实现压 缩比的调整。
( 6 )通过緩存控制器将非恒定压缩比数据进行处理, 使其以 恒定速率输出, 而且通过实时监控緩存的状态及调整压缩比, 以 实现数据的恒定速率输出。 接收端采取相应的解控处理。
实施例二
本发明实施例二提供一种数据压缩方法, 以在 LTE系统 Ir接 口数据压缩方案中的应用为例进行说明, 如图 5 所示的结构。 功 率 /幅度调整器为 AGC, 功率 /幅度补偿器为 AGC补偿。
本实施例中, Ir接口协议支持 BBU和 RRU间所有有效数据 的传输, 最大 IQ数据采样宽度为 16位, 光纤传输需要 8B/10B编 码, 即编码底层传输设备 serdes 所具备的链路编码方式, 目的使 链路传输 0、 1个数均衡, 接收端可以从链路上恢复时钟。
对于 LTE系统上 /下行最大传输速率为:
2 天线情况下, 空口速率: 30.72M*32bit*2 天线 * ( 10/8 ) =2.4576Gbps
4 天线情况下, 空口速率: 30.72M*32bit*4 天线 * ( 10/8 ) =4.9152Gbps
8 天线情况下, 空口速率: 30.72M*32bit*8 天线 * ( 10/8 ) =9.8304Gbps
进一步的, 由于 Ir接口需要考虑支持 8天线, 20M带宽的数 据传输, 需要较高速率的光纤, 因此, 物理层算法需要考虑的问 题为: 如何采取合理可行的算法, 以尽可能小的代价完成大量数 据的可靠有效的传输,即是否有可能进一步降低传输速率,从 16bit 压缩到 M(0<M<16)比特的量级。
因此, 针对 LTE系统信号的特点, 统计上下行信号的分布, 可采用本发明实施例提供的技术方案, 具体实现包括:
步骤 (1 )、 将信号进行 AGC处理, 调整信号的幅度。
步骤(2 )、 存储码本, 码本中的数据位宽为 16bit, 与 AGC之 后的数据位宽一致。
步骤 (3 )、 将信号采用生成的量化码本进行量化。 其中量化 码本根据信号的分布特点训练产生。
步骤(4 )、 将码本下标利用差分编码和分段压缩的方式压缩, 并通过緩存控制经光纤以固定 Mbit传送。
本实施例中, 引入反馈机制, 通过监控緩存的剩余空间的大 小实时的计算压缩比, 并反馈至自适应量化器。 如果当前压缩比 较低时, 可通过几种方式来提高:
1、 改变量化参数, 如采用次优码本;
2、 降低压缩的位宽, 如降为 7bit;
3、 自适应改变分段大小, 如通过一定的步长设置, 实现分段 信息的编码, 建立分段信息的表格, 将分段信息对应的表格区间 下标与位宽信息及数据复用传送。
其中, 位宽信息, 自适应量化信息及分段信息可放置于信号 的循环前缀中, 与数据复用传输。
步骤 (5 )、 在接收端, 先对比特数据进行分段解压缩, 解差 分, 再利用码本通过查表的方式查找对应的数据, 完成解压缩、 解码及 AGC因子恢复。
通过上述方案, 够有效的将 16bit 位宽的数据压缩至 Mbit(0<M<16)。
实施例三
基于与上述方法同样的发明构思, 本发明实施例中还提供了 一种数据压缩设备, 如图 6所示, 该设备包括:
量化模块 11 ,用于根据自适应量化信息对输入信号进行量化; 压缩模块 12, 用于对量化后的输入信号进行压缩;
发送模块 13 , 用于将压缩后的输入信号发送给接收端。
所述自适应量化信息包括以下之一或任意组合: 量化方式; 量化参数; 量化位宽。
所述量化模块 11 , 具体用于当所述自适应量化信息为量化方 式时, 根据所述输入信号的分布特性确定量化方式, 并通过所述 量化方式对所述输入信号进行量化。
所述量化模块 11 ,进一步用于当所述输入信号为均勾分布时, 确定所述量化方式为均匀量化器; 当所述输入信号为非均匀分布 时, 确定所述量化方式为非均匀量化器。
所述量化模块 11 , 具体用于当所述自适应量化信息为量化位 宽时, 监控緩存状态, 并根据所述緩存状态调整量化后的比特位 宽。
所述量化模块 11 , 进一步用于当所述緩存状态低于预设第一 阈值时, 提高所述量化后的比特位宽; 当所述緩存状态高于预设 第二阈值时, 降低所述量化后的比特位宽。
所述量化模块 11 , 具体用于对所述输入信号进行功率或幅度 的调整, 并根据自适应量化信息对调整后的输入信号进行量化。
所述压缩模块 12, 具体用于确定数据段信息, 并利用所述数 据段信息对所述量化后的输入信号进行分段, 在每段内采用相同 的位宽信息对所述量化后的输入信号进行压缩。
所述压缩模块 12, 还用于利用指定编码方式对所述位宽信息 以及所述量化后的输入信号进行编码处理。
所述发送模块 13 , 具体用于对所述压缩后的输入信号进行緩 存, 并以满足预设策略的信号速率将压缩后的输入信号发送给所
述接收端。
其中, 本发明装置的各个模块可以集成于一体, 也可以分离 部署。 上述模块可以合并为一个模块, 也可以进一步拆分成多个 子模块。
通过以上的实施方式的描述, 本领域的技术人员可以清楚地 了解到本发明可借助软件加必需的通用硬件平台的方式来实现, 当然也可以通过硬件, 但很多情况下前者是更佳的实施方式。 基 于这样的理解, 本发明的技术方案本质上或者说对现有技术做出 贡献的部分可以以软件产品的形式体现出来, 该计算机软件产品 存储在一个存储介质中, 包括若干指令用以使得一台计算机设备 (可以是个人计算机, 服务器, 或者网络设备等) 执行本发明各 个实施例所述的方法。
本领域技术人员可以理解附图只是一个优选实施例的示意 图, 附图中的模块或流程并不一定是实施本发明所必须的。
本领域技术人员可以理解实施例中的装置中的模块可以按照 实施例描述进行分布于实施例的装置中, 也可以进行相应变化位 于不同于本实施例的一个或多个装置中。 上述实施例的模块可以 合并为一个模块, 也可以进一步拆分成多个子模块。
上述本发明实施例序号仅仅为了描述, 不代表实施例的优劣。 非局限于此, 任何本领域的技术人员能思之的变化都应落入本发 明的保护范围。
Claims
1、 一种数据压缩方法, 其特征在于, 包括:
发送端根据自适应量化信息对输入信号进行量化;
所述发送端对量化后的输入信号进行压缩;
所述发送端将压缩后的输入信号发送给接收端。
2、 如权利要求 1所述的方法, 其特征在于, 所述自适应量化信 息包括以下之一或任意组合:
量化方式;
量化参数;
量化位宽。
3、 如权利要求 2所述的方法, 其特征在于, 所述发送端根据自 适应量化信息对输入信号进行量化, 包括:
当所述自适应量化信息为量化方式时,所述发送端根据所述输入 信号的分布特性确定量化方式,并通过所述量化方式对所述输入信号 进行量化;
当所述自适应量化信息为量化位宽时, 所述发送端监控緩存状 态, 并根据所述緩存状态调整量化后的比特位宽。
4、 如权利要求 3所述的方法, 其特征在于, 所述发送端根据所 述输入信号的分布特性确定量化方式, 包括:
当所述输入信号为均勾分布时,所述发送端确定所述量化方式为 均匀量化器;
当所述输入信号为非均匀分布时,所述发送端确定所述量化方式 为非均匀量化器。
5、 如权利要求 3所述的方法, 其特征在于, 所述发送端根据所 述緩存状态调整量化后的比特位宽, 包括:
当所述緩存状态低于预设第一阈值时,所述发送端提高所述量化 后的比特位宽;
当所述緩存状态高于预设第二阈值时,所述发送端降低所述量化 后的比特位宽。
6、 如权利要求 1-5任一项所述的方法, 其特征在于, 所述发送 端根据自适应量化信息对输入信号进行量化, 包括:
所述发送端对所述输入信号进行功率或幅度的调整,并根据自适 应量化信息对调整后的输入信号进行量化。
7、 如权利要求 1所述的方法, 其特征在于, 所述发送端对量化 后的输入信号进行压缩, 包括:
所述发送端确定数据段信息,并利用所述数据段信息对所述量化 后的输入信号进行分段,在每段内采用相同的位宽信息对所述量化后 的输入信号进行压缩。
8、 如权利要求 7所述的方法, 其特征在于, 所述方法进一步包 括:
所述发送端利用指定编码方式对所述位宽信息以及所述量化后 的输入信号进行编码处理。
9、 如权利要求 1所述的方法, 其特征在于, 所述发送端将压缩 后的输入信号发送给接收端, 包括:
所述发送端对所述压缩后的输入信号进行緩存,并以满足预设策 略的信号速率将压缩后的输入信号发送给所述接收端。
10、 一种数据压缩设备, 其特征在于, 包括:
量化模块, 用于根据自适应量化信息对输入信号进行量化; 压缩模块, 用于对量化后的输入信号进行压缩;
发送模块, 用于将压缩后的输入信号发送给接收端。
11、 如权利要求 10所述的设备, 其特征在于, 所述自适应量化 信息包括以下之一或任意组合:
量化方式;
量化参数;
量化位宽。
12、 如权利要求 11所述的设备, 其特征在于,
所述量化模块, 具体用于当所述自适应量化信息为量化方式时, 根据所述输入信号的分布特性确定量化方式,并通过所述量化方式对 所述输入信号进行量化;具体用于当所述自适应量化信息为量化位宽 时, 监控緩存状态, 并根据所述緩存状态调整量化后的比特位宽。
13、 如权利要求 12所述的设备, 其特征在于,
所述量化模块, 进一步用于当所述输入信号为均勾分布时, 确定 所述量化方式为均勾量化器; 当所述输入信号为非均勾分布时, 确定 所述量化方式为非均勾量化器。
14、 如权利要求 11所述的设备, 其特征在于,
所述量化模块, 进一步用于当所述緩存状态低于预设第一阈值 时, 提高所述量化后的比特位宽; 当所述緩存状态高于预设第二阈值 时, 降低所述量化后的比特位宽。
15、 如权利要求 11-14任一项所述的设备, 其特征在于, 所述量化模块, 具体用于对所述输入信号进行功率或幅度的调 整, 并根据自适应量化信息对调整后的输入信号进行量化。
16、 如权利要求 10所述的设备, 其特征在于,
所述压缩模块, 具体用于确定数据段信息, 并利用所述数据段信 息对所述量化后的输入信号进行分段,在每段内采用相同的位宽信息 对所述量化后的输入信号进行压缩。
17、 如权利要求 16所述的设备, 其特征在于,
所述压缩模块,还用于利用指定编码方式对所述位宽信息以及所 述量化后的输入信号进行编码处理。
18、 如权利要求 10所述的设备, 其特征在于,
所述发送模块, 具体用于对所述压缩后的输入信号进行緩存, 并 以满足预设策略的信号速率将压缩后的输入信号发送给所述接收端。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110200699.2A CN102291773B (zh) | 2011-07-18 | 2011-07-18 | 一种数据压缩方法和设备 |
CN201110200699.2 | 2011-07-18 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013010437A1 true WO2013010437A1 (zh) | 2013-01-24 |
Family
ID=45337819
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2012/078061 WO2013010437A1 (zh) | 2011-07-18 | 2012-07-02 | 一种数据压缩方法和设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN102291773B (zh) |
WO (1) | WO2013010437A1 (zh) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102291773B (zh) * | 2011-07-18 | 2014-12-10 | 电信科学技术研究院 | 一种数据压缩方法和设备 |
CN102612079B (zh) * | 2012-02-22 | 2014-12-17 | 大唐移动通信设备有限公司 | 一种lte-ir接口数据压缩方法及其装置 |
CN102790656B (zh) * | 2012-05-30 | 2015-10-28 | 新邮通信设备有限公司 | 一种iq数据压缩方法和系统 |
CN103210590B (zh) * | 2012-08-21 | 2016-09-28 | 华为技术有限公司 | 压缩方法及设备 |
CN103117958B (zh) * | 2013-01-08 | 2015-11-25 | 北京百度网讯科技有限公司 | 网络数据包聚集方法、系统及装置 |
CN104207798B (zh) * | 2013-05-31 | 2017-12-22 | 上海联影医疗科技有限公司 | 计算机断层图像的重建方法 |
CN104077505A (zh) * | 2014-07-16 | 2014-10-01 | 苏州博联科技有限公司 | 一种提高16Kbps码率音频数据压缩编码音质方法 |
CN111435346B (zh) * | 2019-01-14 | 2023-12-19 | 阿里巴巴集团控股有限公司 | 离线数据的处理方法、装置及设备 |
CN112085183B (zh) * | 2019-06-12 | 2024-04-02 | 上海寒武纪信息科技有限公司 | 一种神经网络运算方法及装置以及相关产品 |
CN112306974B (zh) * | 2019-07-30 | 2024-10-22 | 深信服科技股份有限公司 | 一种数据处理方法、装置、设备及存储介质 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1285691A (zh) * | 1999-08-20 | 2001-02-28 | 三星电子株式会社 | 根据网络带宽自适应地控制数据传输速率的装置 |
CN1595397A (zh) * | 2004-07-14 | 2005-03-16 | 华南理工大学 | 可听文本的自动制作和播放的方法 |
CN1845099A (zh) * | 2006-04-27 | 2006-10-11 | 掌富科技(南京)有限公司 | 智能移动终端多媒体电子文档转换存储方法及转换存储结构 |
CN102291773A (zh) * | 2011-07-18 | 2011-12-21 | 电信科学技术研究院 | 一种数据压缩方法和设备 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07135652A (ja) * | 1993-11-11 | 1995-05-23 | Kokusai Electric Co Ltd | 静止画データの圧縮方法及び装置、静止画データの伸長方法及び装置 |
JP4304802B2 (ja) * | 1999-12-17 | 2009-07-29 | ソニー株式会社 | ダビング装置 |
CN101715143B (zh) * | 2008-10-07 | 2011-09-14 | 华为技术有限公司 | 自适应量化和自适应反量化的方法、装置及其系统 |
-
2011
- 2011-07-18 CN CN201110200699.2A patent/CN102291773B/zh active Active
-
2012
- 2012-07-02 WO PCT/CN2012/078061 patent/WO2013010437A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1285691A (zh) * | 1999-08-20 | 2001-02-28 | 三星电子株式会社 | 根据网络带宽自适应地控制数据传输速率的装置 |
CN1595397A (zh) * | 2004-07-14 | 2005-03-16 | 华南理工大学 | 可听文本的自动制作和播放的方法 |
CN1845099A (zh) * | 2006-04-27 | 2006-10-11 | 掌富科技(南京)有限公司 | 智能移动终端多媒体电子文档转换存储方法及转换存储结构 |
CN102291773A (zh) * | 2011-07-18 | 2011-12-21 | 电信科学技术研究院 | 一种数据压缩方法和设备 |
Also Published As
Publication number | Publication date |
---|---|
CN102291773A (zh) | 2011-12-21 |
CN102291773B (zh) | 2014-12-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2013010437A1 (zh) | 一种数据压缩方法和设备 | |
Nanba et al. | A new IQ data compression scheme for front-haul link in centralized RAN | |
CN106576393B (zh) | 增强的远程无线电头和计算机可读存储介质 | |
CN111133704B (zh) | 用于基于代码块组的第五代(5g)或其它下一代系统的自适应两级下行链路控制信道结构 | |
US10230394B2 (en) | Methods for compressing and decompressing IQ data, and associated devices | |
WO2014029260A1 (zh) | 数据压缩发送及解压缩方法和设备 | |
WO2012155718A1 (zh) | 一种同相正交数据压缩方法和装置 | |
CN102255692A (zh) | 一种数据压缩方法和设备 | |
CN101754490B (zh) | 一种数据传输的方法、系统和装置 | |
WO2012155614A1 (zh) | 无线通讯系统中数据压缩与解压缩方法、装置及系统 | |
WO2018077163A1 (en) | Quantization-based modulation and coding scheme for mobile fronthaul | |
US20230269023A1 (en) | Radio channel data processing method, communication apparatus, and communication device | |
US20230143476A1 (en) | Methods and systems for reducing fronthaul bandwidth in a wireless communication system | |
CN114157722A (zh) | 一种数据传输方法及装置 | |
CN103138868A (zh) | Lte系统中数据处理方法 | |
Lee et al. | A simple and efficient IQ data compression method based on latency, EVM, and compression ratio analysis | |
CN108092669B (zh) | 一种基于离散余弦变换的自适应数据压缩方法及系统 | |
WO2014129945A1 (en) | Determination of network parameters in mobile communication networks | |
JP2014230098A (ja) | 通信装置及びその制御方法 | |
CN102307372A (zh) | 一种基于Lloyd-Max量化器的数据压缩方法和设备 | |
Feng et al. | Wireless federated learning with dynamic quantization and bandwidth adaptation | |
You | Near-lossless compression/decompression algorithms for digital data transmitted over fronthaul in C-RAN | |
WO2013007185A1 (zh) | 一种数据压缩方法和设备 | |
EP2271019B1 (en) | Mobile communication system using adaptive compression | |
CN103987083B (zh) | 一种视频流的无比率编码解码方法及装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12814334 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12814334 Country of ref document: EP Kind code of ref document: A1 |