CN102104391B - Intermediate-frequency filtering device and method - Google Patents

Abstract

The invention discloses an intermediate frequency filter device and a filter method. The intermediate frequency filter device includes an intermediate frequency filter, including a connected filter coefficient memory and a coefficient selection and control unit; the output end of the coefficient selection and control unit is connected to the intermediate frequency filter. The filter coefficient memory is used to store the filter coefficients of the intermediate frequency filter under different communication systems; the coefficient selection and control unit is used to extract the filter coefficients of the intermediate frequency filter from the filter coefficient memory. The filter coefficients corresponding to the communication standard are provided to the intermediate frequency filter; the intermediate frequency filter is used to perform data filtering processing using the filter coefficients obtained from the coefficient selection and control unit. The intermediate frequency filtering device and filtering method of the present invention can save hardware resources required for digital intermediate frequency information processing in scenarios supporting multi-mode, mixed-mode, multi-antenna and multi-carrier.

Description

An intermediate frequency filtering device and filtering method

technical field

The invention relates to the fields of wireless communication, digital signal processing and integrated circuit design, in particular to an intermediate frequency filter device and a filter method.

Background technique

Software radio technology (Software Radio, referred to as SDR) has become a research hotspot in wireless communication in recent years. Major equipment manufacturers are competing to launch SDR-based systems. The traditional network and base station systems are in a transitional period: 1 , The application of the separated base station architecture has made the base band processing unit (Base Band Unit, BBU for short)-Remote Radio Unit (RRU for short) base station architecture gradually become the mainstream. 2. Miniaturization of base stations has become a development trend. 3. The energy consumption and efficiency requirements of the base station are getting higher and higher, and it has become a key technical indicator for measuring the base station. 4. The application of SDR technology enables smooth transition and evolution of 2G, 3G, and 4G communication technologies, and even mixed-mode base stations of various standards have become an important requirement of operators. Various companies have also established SDR-based soft base station platforms. The BBU-RRU base station architecture is the main base station form in the future. Its development trend is: miniaturization, low cost, low power consumption, and multi-mode compatibility.

With the development of SDR technology, the hardware implementation of digital intermediate frequency processing (FPGA or ASIC implementation) also rises, and with the increase of mixed-mode and multi-mode market demand, the cost, power consumption, and chip area of digital intermediate frequency processing face constraints. The pressure is also increasing. Need to handle more antenna channels, need to support larger bandwidth requirements, need to support multi-mode common-mode mixed-mode processing, all of which will bring double expansion of hardware implementation area.

The intermediate frequency processing of the signal is divided into uplink processing and downlink processing. As shown in Figure 1, the processing process of the downlink is as follows: the baseband processing unit outputs the baseband signal, and in the intermediate frequency processing unit, it undergoes shaping filtering, interpolation filtering, After frequency mixing, peak clipping (Crest Factor Reduction, CFR for short), and digital pre-distortion (Digital Pre-distortion, DPD for short) processing, it is sent to the digital-to-analog conversion device (DAC) to complete the digital-to-analog conversion and then sent to the RF processing unit ; At the same time, the digital predistortion (DPD) processing device also performs predistortion calculation according to the feedback signal of the radio frequency processing unit after analog-to-digital conversion (ADC). As shown in Figure 2, the processing process of the downlink is: the radio frequency processing unit inputs the signal into the intermediate frequency processing unit, after analog-to-digital conversion (ADC), it undergoes frequency mixing processing in the intermediate frequency processing unit, and each carrier is separated and extracted separately. After filtering and shaping filtering, the baseband signal is obtained and sent to the baseband processing unit.

It can be seen from Figure 1 and Figure 2 that in the uplink and downlink of IF processing, the shaping filter, interpolation filter, and decimation filter are key computing units that affect hardware resources. Taking shaping filtering as an example, shaping filters using different communication systems have different coefficients and different filter orders. The usual design method is that each standard corresponds to a shaping filter, and the number of shaping filters corresponding to each standard is determined by the specific number of antennas and carriers. Therefore, in the solution supporting multi-mode and mixed-mode, the number of filters will be very large. The multi-mode mode means that the same set of hardware circuits can support single-mode modes of multiple standards, and the mixed-mode mode means that it can support the simultaneous operation of multiple systems. Assuming that there are N single-mode modes in total, it can support mixed-mode modes of M systems, and M is greater than 1 is an integer less than N, and M generally takes 2 or 3.

In the current implementation scheme, the method of simple replication is often used to support the requirements of more antennas, wider bandwidth, multi-mode, and mixed-mode modes. The simple replication method is implemented as follows: Assuming that a set of intermediate frequency processing is required to support a single antenna When the digital hardware unit supports two antennas, two sets of the same IF processing digital hardware unit are used for realization; assuming that a set of IF processing digital hardware unit is required to support 5M bandwidth, two sets of the same IF processing digital hardware unit are used to support 10M bandwidth Unit implementation; assuming that a set of IF processing digital hardware units is required for single-mode processing, multiple sets of digital hardware units are used to process different modes of IF processing when mixed-mode requirements are supported.

The above-mentioned simple duplication method is simple to implement and easy to expand computing power, but it brings a great waste of hardware resources and chip area, resulting in expansion of base station volume, increase of power consumption, and waste of production cost.

Contents of the invention

The technical problem to be solved by the present invention is to provide an intermediate frequency filtering device and filtering method, which can save hardware resources required for digital intermediate frequency information processing in scenarios supporting multi-mode, mixed-mode, multi-antenna and multi-carrier.

In order to solve the above problems, the present invention provides an intermediate frequency filtering device, including an intermediate frequency filter, including a connected filter coefficient memory and a coefficient selection and control unit; the output end of the coefficient selection and control unit is connected with the intermediate frequency filter; The filter coefficient memory is used to store the filter coefficients of the intermediate frequency filter under different communication systems; the coefficient selection and control unit is used to extract the communication system under the operation of the intermediate frequency filter from the filter coefficient memory The corresponding filter coefficients are provided to the intermediate frequency filter; the intermediate frequency filter is used to perform data filtering processing using the filter coefficients obtained from the coefficient selection and control unit.

Further, the above-mentioned intermediate frequency filter device also has the following characteristics:

The coefficient selection and control unit is used to extract the switched coefficients from the filter coefficient memory when the intermediate frequency filter device operates in a multi-mode or mixed-mode communication mode and when the intermediate frequency filter device switches communication systems. The filter coefficient corresponding to the current communication system is output to the intermediate frequency filter.

Further, the above-mentioned intermediate frequency filter device also has the following characteristics:

The intermediate frequency filter device operates in a multi-mode communication mode or a mixed-mode communication mode and includes N single-mode modes, N is an integer greater than or equal to 1, and if all single-mode modes satisfy Ci*Ai<F/Ni, then the The number of intermediate frequency filters is 1; otherwise, the number of intermediate frequency filters is the result of the maximum value of Ci*Ai<F/Ni corresponding to each single mode and the result of rounding up; among them, Ci is the i-th The number of carriers corresponding to each communication system, Ai is the number of antennas corresponding to the i-th communication system, F is the working clock frequency of the IF filter, Ni is the sampling frequency of the IF filter under the i-th communication system, and i is greater than or equal to 1 and an integer less than or equal to N.

Further, the above-mentioned intermediate frequency filter device also has the following characteristics:

The communication standard refers to the communication standard of the global mobile communication system, the communication standard of the code division multiple access system, the communication standard of the universal mobile communication system, the communication standard of the long term evolution system, and the communication standard of the time division synchronous code division multiple access system.

Further, the above-mentioned intermediate frequency filter device also has the following characteristics:

The intermediate frequency filter is one or a combination of any two or three of the following: shaping filter, interpolation filter, and decimation filter.

In order to solve the above problems, the present invention also provides an intermediate frequency filtering method, including: storing the filtering coefficients of the intermediate frequency filter under different communication systems in the intermediate frequency filtering device; the intermediate frequency filtering device operates in multi-mode or mixed-mode communication mode When down, the filter coefficients corresponding to the communication system under which the intermediate frequency filter works are extracted to perform data filtering processing.

Further, the above intermediate frequency filtering method also has the following characteristics:

The intermediate frequency filter device operates in a multi-mode communication mode or a mixed-mode communication mode, and when the intermediate frequency filter device switches communication systems, extracts filter coefficients corresponding to the switched current communication system for data filtering processing.

Further, the above intermediate frequency filtering method also has the following characteristics:

The intermediate frequency filter device operates in a multi-mode communication mode or a mixed-mode communication mode and includes N single-mode modes, N is an integer greater than or equal to 1, and if all single-mode modes satisfy Ci*Ai<F/Ni, then the The number of intermediate frequency filters is 1; otherwise, the number of intermediate frequency filters is the result of the maximum value of Ci*Ai<F/Ni corresponding to each single mode and the result of rounding up; among them, Ci is the i-th The number of carriers corresponding to each communication system, Ai is the number of antennas corresponding to the i-th communication system, F is the working clock frequency of the IF filter, Ni is the sampling frequency of the IF filter under the i-th communication system, and i is greater than or equal to 1 and an integer less than or equal to N.

Further, the above intermediate frequency filtering method also has the following characteristics:

The communication standard refers to the communication standard of the global mobile communication system, the communication standard of the code division multiple access system, the communication standard of the universal mobile communication system, the communication standard of the long term evolution system, and the communication standard of the time division synchronous code division multiple access system.

Further, the above intermediate frequency filtering method also has the following characteristics:

The intermediate frequency filter is one or a combination of any two or three of the following: shaping filter, interpolation filter, and decimation filter.

The intermediate frequency filtering device and filtering method of the present invention can save hardware resources required for digital intermediate frequency information processing in scenarios supporting multi-mode, mixed-mode, multi-antenna and multi-carrier.

Description of drawings

FIG. 1 is a schematic diagram of a downlink processing method of signal intermediate frequency processing in the prior art;

FIG. 2 is a schematic diagram of an uplink processing method of signal intermediate frequency processing in the prior art;

Fig. 3 is a structural diagram of an intermediate frequency filtering device in an embodiment;

Fig. 4 is a structural diagram of a filter operation unit.

Detailed ways

As shown in Fig. 3, the intermediate frequency filter device includes an intermediate frequency filter, and also includes a filter coefficient memory connected to it and a coefficient selection and control unit; the output end of the coefficient selection and control unit is connected to the intermediate frequency filter. This IF filter is one or a combination of any two or three of the following: shaping filter, interpolation filter, decimation filter;

The filter coefficient memory is used to store the filter coefficients of the IF filter under different communication systems. The filter coefficient memory can be a single memory device, or a combination of multiple sub-memory devices. For example, when the IF filter is a shaping filter, there may be multiple sub-memory devices, and each sub-memory device stores a filter coefficient of a shaping filter of a communication standard. When the intermediate frequency filter is a shaping filter and an interpolation filter, each sub-memory device stores the filter coefficients of the shaping filter and the interpolation filter under a communication system, or each sub-memory device stores a filter coefficient of a communication system. filter coefficients of the filter.

The coefficient selection and control unit is used to extract the filter coefficient corresponding to the communication system under which the intermediate frequency filter works from the filter coefficient memory and provide it to the intermediate frequency filter. Specifically, when the intermediate frequency filter device operates in a multi-mode communication mode or a mixed-mode communication mode, when the intermediate frequency filter device switches the communication system, it extracts the filter coefficient corresponding to the switched current communication system from the filter coefficient memory, and Output to IF filter. When the filter coefficient memory is a combination of multiple sub-memory devices, the coefficient selection and control unit extracts the filter coefficients corresponding to the communication system under which the intermediate frequency filter works from the corresponding sub-memory devices.

An intermediate frequency filter is used to perform data filtering processing using the filter coefficients obtained from the coefficient selection and control unit.

The structure of the IF filter is shown in Figure 4, which mainly includes a delay unit Z ^-1 , a multiplier and an adder. a ₀ , a ₁ ,..., a _n are filter coefficients, which are input from the filter coefficient memory, x(n) is input data, and y(n) is output data.

Using the way of storing filter coefficients in advance can realize the common use of the same hardware for various standard single-mode filters. Different systems can be configured with different coefficients to achieve their own support. The coefficient configuration method is to write each The filter coefficients corresponding to the standard, this method avoids the waste of hardware resources caused by the use of multiple different filters for different standard filters. Using filter time division multiplexing technology can further save hardware resources.

When the IF filter works in the multi-mode communication mode, the working clock frequency of the IF filter is set to F. Under this multi-mode communication mode, there are N kinds of single-mode modes in total, and the input sampling rates are N1, N2,..., Nn. Then, the multiplexing ratios of the time-division multiplexing processing of the filter corresponding to various standards are respectively F/N1, F/N2, . . . , F/Nn. It is assumed that the numbers of carriers corresponding to the N systems are respectively C1, C2, . . . , and the numbers of Cn antennas are respectively A1, A2, . . . , An. Then, the processing capacity requirements corresponding to each standard are respectively C1*A1, C2*A2, . . . , Cn*An. If C1*A1<F/N1, C2*A2<F/N2, . . . , Cn*An<F/Nn are simultaneously satisfied. Then support all single-mode modes and only need one IF filter to achieve through time division multiplexing, otherwise, the number of IF filters is C1*A1*N1/F, C2*A2*N2/F, .. ., the result of rounding up the maximum value Ci*Ai*Ni/F (i is any value between 1 and n) in Cn*An*Nn/F.

The intermediate frequency filter can also work in the mixed-mode communication mode. The premise of the mixed-mode processing is that compared with the single-mode processing, the total processing bandwidth of the hardware remains unchanged, that is, the bandwidth originally enjoyed by the single-mode processing is now divided by several mixed-mode The communication methods are shared, so the total processing capacity of the hardware remains unchanged. Time division multiplexing in the mixed-mode mode is much more complicated than that in the single-mode mode, because it involves time-division multiplexing of several systems. If the same filter is used to implement, the filter coefficients need to be switched. The current system is processed in time-division multiplexing. At the end of the process, it is necessary to switch to the filter coefficients of the next system for calculation, so the filter structure shown in Figure 3 is adopted, and M coefficient memories are set to correspond to M different systems in the mixed-mode mode, which are selected and controlled by the coefficients The unit is responsible for switching the coefficient memory and controlling the calculation process of the filter during the time division multiplexing process. Using the filter extraction sharing technology to avoid using M filters to process the filtering of the mixed-mode communication mode in M separately is equivalent to extracting the coefficient memories of the M filters independently. In the calculation unit part of the M filters, only One of the filter operation units is reserved (when only one filter is needed to time-division multiplex the mixed-mode communication mode), and the M coefficient memories are reserved, and the different systems are realized by switching the coefficient memories during the time-division multiplexing process. filter processing. Thereby saving hardware resources.

In the mixed-mode communication mode of M standards, a filter at a certain stage only needs M coefficient memories and ceil(Ci*Ai*Ni/F) filter operation units at most, and ceil is a ceiling function.

In the above method, the filter coefficient memory extraction and sharing technology can effectively improve the filter multiplexing degree in the mixed-mode mode, and greatly reduce the consumption of hardware resources in the mixed-mode mode.

The single-mode communication system supported in the multi-mode communication mode or mixed-mode communication mode in the present invention includes the following systems: Global System for Mobile Communications (Global System for Mobile Communications, referred to as GSM) communication system, Code-Division Multiple Access system (Code-Division Multiple Access (CDMA for short) communication standard, Universal Mobile Telecommunications System (UMTS for short) communication standard, Long Term Evolution (LTE for short) 1.4M communication standard, LTE3M communication standard, LTE5M communication standard, LTE10M communication Standard, LTE15M communication standard, LTE20M communication standard, Time Division-Synchronous Code Division Multiple Access (Time Division-Synchronous Code Division Multiple Access, referred to as TD-SCDMA) communication standard.

In order to more clearly illustrate the working principle of the method and hardware device proposed by the present invention, the following description will be made in conjunction with specific embodiments.

The shape filter design is taken as an example for illustration, and other filter design methods are similar.

According to the requirements, the input sampling rate and filter usage of each standard forming filter are shown in the following table:

standard input sampling rate Number of Antenna Carriers (Downlink) Single-mode multiplexing ratio 245.76M clock number of filters GSM 1.28 4*8 96 1 TD-SCDMA 1.28 4*12 96 1 LTE1.4M 1.92 4*8 64 1

CDMA 1.92 4*8 64 1 UMTS 3.84 4*4 32 1 LTE3M 3.84 4*4 32 1 LTE5M 7.68 4*4 16 1 LTE10M 15.36 4*2 8 1 LTE15M 30.72 4*1 4 1 LTE20M 30.72 4*1 4 1

In the above table, the number of antenna carriers is the maximum number of carriers supported by 4 antennas in single-mode mode. The number of antennas in the table is unified as 4. The single-mode multiplexing ratio is calculated according to the circuit working clock of 245.76M, and the input data of each system is I , Q multiplexing data. It can be seen from the table that the single-mode multiplexing ratio of each standard is greater than or equal to the number of antenna carriers to be processed, so the filter time division multiplexing technology can use one shaping filter to support the single-mode shaping filtering of each standard. In the mixed-mode mode, the total processing bandwidth remains unchanged. For example, the mixed-mode support for GUL three standards is: GSM standard 2 carriers/antenna, UMTS standard 1 carrier/antenna, and LTE10M standard 1 carrier/antenna. When the three systems are mixed, one shaping filter can still be used for time division multiplexing processing, but at this time, three coefficient memories are required to store the shaping filter coefficients corresponding to GSM, the shaping filter coefficients corresponding to UMTS and the corresponding LTE10M Shaping filter coefficients. Because the input sampling rate of the GSM system is the lowest among the three systems, the multiplexing ratio calculated by GSM is 96, and the processing time of the entire filter is divided into 96 equal time segments, and the single-antenna 2-carrier GSM processing needs to occupy 2 of them. 4 antennas occupy 8 segments; the input sampling rate of UMTS is 3 times that of GSM, and it takes 3 time segments to process a single antenna and single carrier, so it takes 12 time segments to process 4 antennas; the input sampling rate of LTE10M is Twelve times that of GSM, it takes 12 time segments to process a single antenna and a single carrier, and a total of 48 time segments for 4 antennas, so the total time division multiplexing processing of the three systems takes 68 time segments, and there are still 28 time segments Processing time available. During the mixed-mode time-division multiplexing process of the three systems, at the beginning of the first time segment, the filter coefficients are read from the coefficient memory 1, and the coefficient memory 1 stores the filter coefficients corresponding to the GSM standard. At the end of the segment, the filter calculation is completed. At this time, the coefficient is switched to input from the coefficient memory 2. The coefficient memory 2 stores the filter coefficient corresponding to the UMTS system. When the UMTS filter calculation is completed at the end of the 20th time segment, the coefficient is switched to Input from the coefficient memory 3, and the coefficient memory 3 stores filter coefficients corresponding to the LTE10M standard. The filter calculation is completed until the end of the 68th time segment. During the whole time division multiplexing process, by switching the coefficient memory, one filter is used to complete the shaping filtering for the three systems. Save hardware resource consumption. The number of coefficient memories is determined by the maximum number of mixed-mode communication systems that need to be supported, and the coefficients corresponding to each system are pre-written into the coefficient memory according to the mixed-mode configuration. In the single-mode mode, only one of the coefficient memories needs to be used, and according to the single-mode configuration, the filter coefficients corresponding to the single-mode mode are written into the coefficient memory in advance.

The method proposed by the invention can realize the maximum optimization of hardware resources by combining technologies such as filter time-division multiplexing, filter coefficient configurability, and filter coefficient memory extraction and multiplexing.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (6)

1. An intermediate frequency filtering device, comprising an intermediate frequency filter, is characterized in that, It includes a connected filter coefficient memory and a coefficient selection and control unit; the output end of the coefficient selection and control unit is connected to the intermediate frequency filter; The filter coefficient memory is used to store the filter coefficients of the intermediate frequency filter under different communication systems; The coefficient selection and control unit is used to extract the filter coefficients corresponding to the communication system under which the intermediate frequency filter works from the filter coefficient memory and provide them to the intermediate frequency filter, including: the intermediate frequency filter device Operating in the multi-mode or mixed-mode communication mode, when the intermediate frequency filter device switches the communication system, extracts the filter coefficient corresponding to the switched current communication system from the filter coefficient memory, and outputs it to the intermediate frequency filter ; The multimode communication mode or the mixed mode communication mode of the intermediate frequency filtering device includes N kinds of single mode modes in total, and N is an integer greater than or equal to 1. If Ci*Ai<F/Ni is satisfied under each single mode mode, then all The number of intermediate frequency filters is 1; otherwise, the number of intermediate frequency filters is the result of the maximum value in the value of Ci*Ai*Ni/F corresponding to each single mode and the result of rounding up; wherein, Ci is the first The number of carriers corresponding to the i-th communication system, Ai is the number of antennas corresponding to the i-th communication system, F is the working clock frequency of the IF filter, Ni is the sampling frequency of the IF filter under the i-th communication system, and i is greater than or equal to 1 and an integer less than or equal to N; The intermediate frequency filter is used to perform data filtering using the filter coefficients obtained from the coefficient selection and control unit. 2. The intermediate frequency filtering device as claimed in claim 1, characterized in that, The communication standard refers to the communication standard of the global mobile communication system, the communication standard of the code division multiple access system, the communication standard of the universal mobile communication system, the communication standard of the long term evolution system, and the communication standard of the time division synchronous code division multiple access system. 3. The intermediate frequency filter device as claimed in claim 1, characterized in that, The intermediate frequency filter is one or a combination of any two or three of the following: shaping filter, interpolation filter, and decimation filter. 4. A method for intermediate frequency filtering, characterized in that, The filter coefficients of the intermediate frequency filter under different communication systems are stored in the intermediate frequency filter device; when the intermediate frequency filter device operates in a multi-mode or mixed-mode communication mode, the filter corresponding to the communication system under the operation of the intermediate frequency filter is extracted The coefficients are used for data filtering processing, including: the intermediate frequency filtering device operates in a multi-mode communication mode or a mixed-mode communication mode, and when the intermediate frequency filtering device performs communication system switching, extracting filter coefficients corresponding to the switched current communication system Perform data filtering; The multi-mode communication mode or the mixed-mode communication mode of the intermediate frequency filtering device includes N kinds of single-mode communication modes, and N is an integer greater than or equal to 1. If Ci*Ai<F/Ni is satisfied under each single-mode mode, then the The number of intermediate frequency filters is 1; otherwise, the number of intermediate frequency filters is the result of the maximum value of the values of Ci*Ai*Ni/F corresponding to each single mode and the result of rounding up; where Ci is the i-th The number of carriers corresponding to each communication system, Ai is the number of antennas corresponding to the i-th communication system, F is the working clock frequency of the IF filter, Ni is the sampling frequency of the IF filter under the i-th communication system, and i is greater than or equal to 1 and an integer less than or equal to N. 5. intermediate frequency filtering method as claimed in claim 4, is characterized in that, The communication standard refers to the communication standard of the global mobile communication system, the communication standard of the code division multiple access system, the communication standard of the universal mobile communication system, the communication standard of the long term evolution system, and the communication standard of the time division synchronous code division multiple access system. 6. intermediate frequency filtering method as claimed in claim 4, is characterized in that, The intermediate frequency filter is one or a combination of any two or three of the following: shaping filter, interpolation filter, and decimation filter.