CN101739231A - Booth-Wallace tree multiplier - Google Patents
Booth-Wallace tree multiplier Download PDFInfo
- Publication number
- CN101739231A CN101739231A CN201010013538A CN201010013538A CN101739231A CN 101739231 A CN101739231 A CN 101739231A CN 201010013538 A CN201010013538 A CN 201010013538A CN 201010013538 A CN201010013538 A CN 201010013538A CN 101739231 A CN101739231 A CN 101739231A
- Authority
- CN
- China
- Prior art keywords
- partial product
- circuit
- partial
- multiplier
- intermediate state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Complex Calculations (AREA)
Abstract
本发明公开了一种布斯-华莱士树型乘法器,主要解决现有乘法器延时较大,电路和版图实现复杂的问题。包括布斯编码电路、部分积中间态产生电路和部分积结果产生电路。布斯编码电路对乘数重新编码产生部分积倍数控制信号和符号控制信号,分别输出到部分积中间态产生电路和部分积结果产生电路;部分积中间态产生电路产生相应倍数的部分积中间态,并输出到部分积结果产生电路;部分积结果产生电路,依据部分积控制信号及部分积中间态电路产生结果产生最终的正、负部分积。本发明对部分积产生单元电路进行改进,减少了部分积压缩器级数及数目,有效提高了乘法器的运行速度,并减低电路及版图实现复杂度,可用于各类微处理器、数字信号处理器中。
The invention discloses a Booth-Wallace tree multiplier, which mainly solves the problems that the existing multiplier has a large time delay and complicated circuit and layout implementation. Including Booth encoding circuit, partial product intermediate state generating circuit and partial product result generating circuit. The Booth encoding circuit re-encodes the multiplier to generate a partial product multiple control signal and a sign control signal, which are respectively output to the partial product intermediate state generation circuit and the partial product result generation circuit; the partial product intermediate state generation circuit generates the partial product intermediate state of the corresponding multiple , and output to the partial product result generating circuit; the partial product result generating circuit generates the final positive and negative partial products according to the partial product control signal and the partial product intermediate state circuit generating result. The invention improves the partial product generating unit circuit, reduces the series and number of partial product compressors, effectively improves the operating speed of the multiplier, and reduces the complexity of circuit and layout implementation, and can be used in various types of microprocessors, digital signals in the processor.
Description
技术领域technical field
本发明属于电子元件技术领域,涉及一种乘法器,可用于各类微处理器、数字信号处理器(DSP)中。The invention belongs to the technical field of electronic components and relates to a multiplier which can be used in various microprocessors and digital signal processors (DSP).
背景技术Background technique
乘法器作为基本的运算单元应用于各类微处理器、数字信号处理器(DSP)及专用集成电路(ASIC)中。目前的乘法器已经具有了较高的性能,但由于现代信息处理中安全保密应用重要性的不断提高,对乘法器性能也在不断提出越来越高的要求。As a basic arithmetic unit, the multiplier is used in various microprocessors, digital signal processors (DSP) and application-specific integrated circuits (ASIC). The current multiplier already has high performance, but due to the increasing importance of security and confidentiality applications in modern information processing, the performance of the multiplier is also continuously put forward higher and higher requirements.
早期乘法器大多采用移位加算法及阵列乘法器算法。这两种算法从实现方法来看,乘数的每一位都会产生部分积,且只产生1或0倍的部分积形式。因而,当乘法操作数位宽比较宽时,运算量就非常大。实际上当乘数的某一位为0时,并不会产生具有实际意义的部分积,更不需要进行累加运算。在实际运算的过程中,部分积的产生和部分积累加次数是由乘数中值为“1”的个数决定。布斯编码算法是针对补码运算的一种部分积产生算法,它将乘数重新编码以减少乘数中值为“1”的个数,从而减少了部分积的数目和部分积累加的次数,有效的提高了乘法器运算的效率。在布斯编码算法中,被广泛使用的一种布斯编码算法是改进的布斯编码算法,即二阶布斯编码算法。Most of the early multipliers use shift-add algorithm and array multiplier algorithm. From the perspective of the implementation methods of these two algorithms, each bit of the multiplier will generate a partial product, and only 1 or 0 times the partial product form. Therefore, when the bit width of the multiplication operand is relatively wide, the amount of calculation is very large. In fact, when a certain bit of the multiplier is 0, no partial product with practical significance will be generated, let alone an accumulation operation. In the actual operation process, the generation of partial products and the number of partial accumulations are determined by the number of "1" in the multiplier. The Booth encoding algorithm is a partial product generation algorithm for complement operation, which recodes the multiplier to reduce the number of "1" in the multiplier, thereby reducing the number of partial products and the number of partial accumulations , effectively improving the efficiency of the multiplier operation. In the Booth coding algorithm, a widely used Booth coding algorithm is the improved Booth coding algorithm, that is, the second-order Booth coding algorithm.
改进的布斯编码算法可应用到有符号和无符号的乘法数,它基本上可以将部分积的数目缩减至非布斯编码算法的一半。当乘法操作数N为奇数时,部分积的数目为(N+1)/2,当N为偶数时,部分积的数目为(N/2)+1。它极大程度的减少了部分积压缩模块所占用的面积和产生的延时,提高了乘法器单元的性能并减少了面积。The improved Booth coding algorithm can be applied to signed and unsigned multiplication numbers, and it can basically reduce the number of partial products to half that of the non-Booth coding algorithm. When the multiplication operand N is an odd number, the number of partial products is (N+1)/2, and when N is an even number, the number of partial products is (N/2)+1. It greatly reduces the area occupied by the partial product compression module and the delay generated, improves the performance of the multiplier unit and reduces the area.
在部分积压缩阶段被广泛采用的一种压缩技术是华莱士树压缩算法,它将部分积按列分组,每一列对应一组加法器,各列同时相加,前一列进位传至后一列,生成新的部分积。采用同样方法化简新的部分积,直至剩下最后两行部分积,再采用一种快速加法器相加得积。这种部分积压缩方法通过尽可能的对操作数进行并行运算,以此来达到提高乘法运算速度的目的。所以,布斯编码算法和华莱士压缩树相结合的混合型算法经常被应用于高速乘法器结构中。A compression technique widely used in the partial product compression stage is the Wallace tree compression algorithm, which groups partial products by columns, each column corresponds to a set of adders, and each column is added at the same time, and the carry of the previous column is passed to the next column , generating a new partial product. Use the same method to simplify the new partial products until the last two rows of partial products are left, and then use a fast adder to add the products. This partial product compression method achieves the purpose of improving the speed of multiplication by performing parallel operations on the operands as much as possible. Therefore, the hybrid algorithm combining the Booth coding algorithm and the Wallace compression tree is often used in the high-speed multiplier structure.
在采用这种混合型算法实现乘法器时,改进的布斯算法将乘数按每一组三位进行划分。三位分别是低位、本位、高位,其中低位来自于低一组中的最高位。从而产生{±2,±1,0}倍数的部分积,之后再用华莱士树将这些部分积相加总。When using this hybrid algorithm to implement a multiplier, the improved Booth algorithm divides the multiplier into groups of three bits. The three bits are respectively the low bit, the standard bit, and the high bit, wherein the low bit comes from the highest bit in the low group. Thus, partial products of multiples of {±2, ±1, 0} are generated, and then the Wallace tree is used to sum these partial products.
负部分积的实现分为两个阶段。第一阶段,利用布斯编码算法产生正部分积的非,并输出到华莱士压缩树中。第二阶段,在这一行部分积非的最低有效位加1,这个1放在下一行部分积与本部分积最低有效位相对齐的地方,如图10所示。The realization of the negative part product is divided into two stages. In the first stage, the Booth coding algorithm is used to generate the negative of the positive partial product and output to the Wallace compression tree. In the second stage, 1 is added to the least significant bit of the partial product in this row, and this 1 is placed in the place where the partial product of the next row is aligned with the least significant bit of this partial product, as shown in Figure 10.
由于改进布斯编码算法所产生的部分积有正负之分,压缩时华莱士树结构就会产生很大差异。图6和图8分别为进行了部分积符号扩展的正部分积阵列结构和负部分积阵列结构,图7和图9分别为正部分积压缩器阵列结构和负部分积压缩器阵列结构。其中负部分积阵列结构会因为最后一行部分积符号位而导致多余一行部分的影响。在对这一行部分积进行压缩时就需要额外的压缩器,这一行压缩器,不但增加了压缩级数使得乘法器延时较大,而且在版图实现时这种华莱士树结构也比较复杂。Since the partial product generated by the improved Booth coding algorithm has positive and negative points, the Wallace tree structure will have a great difference when compressed. Figure 6 and Figure 8 respectively show the positive partial product array structure and the negative partial product array structure with partial product sign extension, and Figure 7 and Figure 9 respectively show the positive partial product compressor array structure and the negative partial product compressor array structure. The negative partial product array structure will cause the impact of an extra row due to the sign bit of the last row of partial product. An additional compressor is required when compressing the partial product of this row. This row of compressors not only increases the number of compression stages and makes the delay of the multiplier larger, but also the Wallace tree structure is more complicated when the layout is realized. .
随着人们对高性能乘法器的追求,改进的布斯编码算法由于其产生的部分积数目仍然比较多,不在适合于更高性能的乘法器。产生的部分积数目越少,乘法迭代的次数就越少,从而乘法运算的速度就越快。三阶布斯编码算法产生的部分积是非布斯编码算法的三分之一,所以其越来越多的应用于高性能乘法器中。With people's pursuit of high-performance multipliers, the improved Booth coding algorithm is not suitable for higher-performance multipliers because the number of partial products it produces is still relatively large. The fewer partial products produced, the fewer iterations of the multiplication, and thus the faster the multiplication operation. The partial product generated by the third-order Booth encoding algorithm is one-third of that of the non-Booth encoding algorithm, so it is more and more used in high-performance multipliers.
三阶布斯编码算法将乘数分为四位一组,其中三位来自本组,最低位来自相邻低位组中的最高位,如图13所示。三阶布斯编码算法产生{±4,±3,±2,±1,0}倍数的部分积。±3倍的部分积,常被人们称为“难倍数”部分积,实现时需要在部分积编码模块中增设一个“难倍数”部分积实现电路,图11所示。但这种三阶布斯编码算法与改进的布斯编码算法一样,均会产生负部分积,同样也会导致乘法器延时较大,电路和版图实现复杂的问题。The third-order Booth encoding algorithm divides the multiplier into groups of four bits, three of which come from this group, and the lowest bit comes from the highest bit in the adjacent low-order group, as shown in Figure 13. The third-order Booth encoding algorithm produces partial products of multiples of {±4, ±3, ±2, ±1, 0}. The partial product of ±3 times is often referred to as a "difficult multiple" partial product. When realizing it, it is necessary to add a "difficult multiple" partial product realization circuit in the partial product encoding module, as shown in Figure 11. However, this third-order Booth coding algorithm, like the improved Booth coding algorithm, will generate negative partial products, which will also lead to large multiplier delays and complex circuit and layout implementations.
发明内容Contents of the invention
本发明的目的在于克服上述已有技术的不足,提出一种布斯-华莱士树型乘法器,通过对部分积实现电路的改进,避免因负部分积符号而需要额外部分积压缩器的影响,减少部分积压缩级数,提高乘法器速度,同时使得部分积压缩阵列结构更加规整,易于版图实现。The purpose of the present invention is to overcome above-mentioned deficiencies in the prior art, propose a kind of Booth-Wallace tree type multiplier, realize the improvement of circuit by partial product, avoid needing extra partial product compressor because of negative partial product sign Influence, reduce the number of partial product compression series, increase the speed of the multiplier, and at the same time make the partial product compression array structure more regular, easy to realize the layout.
为实现上述目的,本发明的布斯-华莱士树型乘法器包括:部分积产生模块、部分积压缩模块和最后两行部分积相加总的加法器模块,其中部分积产生模块中改进的部分积产生单元包括:In order to achieve the above object, the Booth-Wallace tree multiplier of the present invention includes: a partial product generation module, a partial product compression module and an adder module of the last two row partial product additions, wherein the partial product generation module improves The partial product generating units include:
布斯编码电路,用于对乘数重新编码以产生部分积的倍数控制信号和符号控制信号。部分积倍数控制信号输出到部分积中间态产生电路,部分积符号控制信号输出到部分积结果产生电路中。A Booth encoding circuit for re-encoding the multiplier to generate a partial product multiple control signal and a sign control signal. The partial product multiple control signal is output to the partial product intermediate state generating circuit, and the partial product sign control signal is output to the partial product result generating circuit.
部分积中间态产生电路,根据布斯编码所产生的控制信号产生相应倍数的部分积中间态,并将部分积中间态输出到部分积结果产生电路中。The partial product intermediate state generation circuit generates partial product intermediate states of corresponding multiples according to the control signal generated by the Booth code, and outputs the partial product intermediate state to the partial product result generating circuit.
部分积结果产生电路,依据布斯编码所产生的控制信号及部分积中间态电路的产生结果产生最终的正、负部分积。The partial product result generating circuit generates the final positive and negative partial products according to the control signal generated by the Booth code and the result of the partial product intermediate state circuit.
所述的部分积结果产生电路由负部分积运算电路和二选一选择器构成,该负部分积运算电路用于对部分积中间态结果进行运算,产生负部分积输出到二选一选择器中;根据布斯编码电路所产生的部分积符号控制信号,该二选一选择器对负部分积运算电路结果和部分积中间态产生电路结果进行选择,产生相应倍数及类型的部分积,并输出到部分积压缩模块中。The partial product result generating circuit is composed of a negative partial product operation circuit and a two-choice selector, and the negative partial product operation circuit is used to perform operations on the partial product intermediate state result, and generates a negative partial product output to the two-choice selector middle; according to the partial product sign control signal generated by the Booth encoding circuit, the two-choice selector selects the result of the negative partial product operation circuit and the partial product intermediate state generation circuit result, and generates a partial product of the corresponding multiple and type, and Output to the partial product compression module.
所述的部分积压缩电路由六级压缩电路组成的阵列结构,该阵列结构形状为规则的直角梯形结构。The partial product compression circuit is an array structure composed of six stages of compression circuits, and the shape of the array structure is a regular right-angled trapezoidal structure.
本发明通过在部分积产生电路中增设负部分积运算电路,改变了最后一行部分积的类型,能够有效减少部分积压缩阵列级数,并使其结构规整化易于版图实现。同时有效提升了乘法器的运行速度,在不增加电路面积的情况下,将16位乘法器的运行速度提升13%。The present invention changes the type of the last row of partial products by adding a negative partial product operation circuit in the partial product generation circuit, can effectively reduce the number of partial product compression arrays, and makes the structure regular and easy to realize in layout. At the same time, the operating speed of the multiplier is effectively improved, and the operating speed of the 16-bit multiplier is increased by 13% without increasing the circuit area.
附图说明Description of drawings
图1为本发明乘法器的整体结构图;Fig. 1 is the overall structural diagram of multiplier of the present invention;
图2为本发明中的最后一个部分积产生电路图;Fig. 2 is that last partial product among the present invention produces circuit diagram;
图3为本发明中的部分积阵列图;Fig. 3 is a partial product array diagram among the present invention;
图4为本发明中的部分积压缩器阵列图;Fig. 4 is a partial product compressor array diagram among the present invention;
图5为传统乘法器的整体结构图;Fig. 5 is the overall structural diagram of traditional multiplier;
图6为传统乘法器中正部分积阵列图;Fig. 6 is positive partial product array diagram in traditional multiplier;
图7为传统乘法器中正部分积压缩器阵列图;Fig. 7 is the positive partial product compressor array diagram in the traditional multiplier;
图8为传统乘法器中负部分积阵列图;Fig. 8 is negative partial product array diagram in the traditional multiplier;
图9为传统乘法器中负部分积压缩器阵列图;Fig. 9 is negative partial product compressor array figure in traditional multiplier;
图10为传统乘法器中负部分积实现图;Fig. 10 is the realization figure of negative partial product in traditional multiplier;
图11为传统乘法器中部分积产生电路图;Fig. 11 is the partial product generation circuit diagram in the traditional multiplier;
图12为传统乘法器中部分积压缩器阵列图;Fig. 12 is an array diagram of a partial product compressor in a traditional multiplier;
图13为三阶布斯编码图。Figure 13 is a diagram of a third-order Booth encoding.
具体的实现方式specific implementation
参照图5,目前公知的三阶布斯-华莱士树型乘法器基本结构主要由部分积的产生模块、部分积压缩模块及最终加法器模块构成,其中:Referring to Fig. 5, the basic structure of currently known third-order Booth-Wallace tree multiplier is mainly composed of a partial product generation module, a partial product compression module and a final adder module, wherein:
部分积产生模块,通过对乘数重新编码产生相应倍数的部分积中间态及部分积符号,并将所产生的部分积中间态及部分积符号输出到部分积压缩模块中。The partial product generation module generates partial product intermediate states and partial product symbols of corresponding multiples by recoding the multipliers, and outputs the generated partial product intermediate states and partial product symbols to the partial product compression module.
部分积压缩阵列模块,依据部分积压缩器阵列结构,对所有部分积中间态和部分积符号同时进行压缩处理,直至产生最后两行部分积,并将其输出到最终加法器模块中。The partial product compression array module, according to the partial product compressor array structure, compresses all partial product intermediate states and partial product symbols at the same time, until the last two rows of partial products are generated, and output them to the final adder module.
最终加法器模块,将部分积压缩阵列所产生的最后两行部分积进行相加总得到乘积结果。The final adder module adds up the partial products of the last two rows generated by the partial product compressed array to obtain the product result.
所述的部分积产生模块,由6个部分积产生单元021构成,其结构如图11所示,它包括:布斯编码电路、一般倍数部分积实现电路和“难倍数”部分积实现电路。该布斯编码电路用于对乘数重新编码,以产生部分积的倍数控制信号和符号控制信号;该一般倍数部分积实现电路主要实现{±4,±2,±1}倍数的部分积中间态;该“难倍数”部分积实现电路主要实现±3倍数的部分积中间态结果。The partial product generating module is composed of 6 partial
布斯编码所产生的部分积倍数控制信号输出到一般倍数部分积实现电路和“难倍数”部分积实现电路,而部分积符号则输出到部分积压缩阵列中。一般倍数实现电路和“难倍数”实现电路将其所产生的部分积中间态结果输出到部分积压缩模块中。The partial product multiplier control signal generated by Booth encoding is output to the general multiple partial product realization circuit and the "difficult multiple" partial product realization circuit, and the partial product symbol is output to the partial product compression array. The general multiple realization circuit and the "difficult multiple" realization circuit output the partial product intermediate state results to the partial product compression module.
传统乘法器中为了简化电路结构,将负部分积的实现分成如图10所示的两个阶段:第一个阶段在部分积产生电路中,产生相应倍数的部分积中间态,并输出到部分积压缩阵列中。第二阶段在部分积压缩阶段,通过在部分积中间态末尾加1产生负部分积。由于部分积产生电路会产生正、负部分积,而正、负部分积阵列结构不同,所以其压缩器阵列结构也将会产生差异。当所有部分积均为正时,部分积符号全为0,所以部分积阵列最后一行不会出现由符号位构成的额外一行部分积,其部分积阵列结构如图6所示,相应的部分积压缩器阵列结构如图7所示。当所有部分积均为负时,部分积符号全为1,其部分积阵列结构最后一行将会出现由符号位组成的部分积,其结构如图8,而部分积压缩器阵列如图9所示。In order to simplify the circuit structure in the traditional multiplier, the realization of the negative partial product is divided into two stages as shown in Figure 10: the first stage is in the partial product generation circuit, which generates the intermediate state of the partial product of the corresponding multiple, and outputs it to the partial in a packed array. The second stage is in the partial product compression stage, by adding 1 to the end of the partial product intermediate state to generate a negative partial product. Since the partial product generation circuit will generate positive and negative partial products, and the array structures of the positive and negative partial products are different, the structure of the compressor array will also be different. When all partial products are positive, the symbols of the partial products are all 0, so the last row of the partial product array will not appear an extra row of partial products composed of sign bits. The structure of the partial product array is shown in Figure 6, and the corresponding partial product The compressor array structure is shown in Figure 7. When all the partial products are negative, the signs of the partial products are all 1, and the last row of the partial product array structure will have a partial product composed of sign bits, as shown in Figure 8, and the partial product compressor array is shown in Figure 9 Show.
由于布斯-华莱士树型乘法器同时需要对正、负部分积进行处理,所以传统的布斯-华莱士树型乘法器结构必须同时兼顾正、负部分积压缩器阵列结构。而负部分积压缩器阵列结构由于需要对符号位压缩而需要额外部分积压缩器,会导致部分积压缩器阵列结构不规整及压缩级数较大的问题,所以传统布斯-华莱士树型乘法器中部分积压缩器阵列结构如图12所示,该结构也存在部分积压缩器阵列结构不规整及压缩级数较大的缺点。Since the Booth-Wallace tree multiplier needs to process positive and negative partial products at the same time, the traditional Booth-Wallace tree multiplier structure must take into account both positive and negative partial product compressor array structures. However, the negative partial product compressor array structure needs an additional partial product compressor due to the need to compress the sign bit, which will lead to the problem of irregular partial product compressor array structure and large number of compression stages. Therefore, the traditional Booth-Wallace tree The structure of the partial product compressor array in the type multiplier is shown in Fig. 12. This structure also has the disadvantages of irregular structure of the partial product compressor array and large number of compression stages.
为了减少传统布斯-华莱士树型乘法器中延时较大,部分积压缩器阵列结构复杂的问题,本发明通过在最后一行部分积产生电路中增设负部分积运算电路,降低了传统布斯-华莱士树型乘法器延时,解决了部分积压缩器阵列结构复杂,版图实现难的问题。In order to reduce the problem of large delay and complex array structure of partial product compressors in the traditional Booth-Wallace tree multiplier, the present invention reduces the traditional The delay of the Booth-Wallace tree multiplier solves the problem that the array structure of the partial product compressor is complex and the layout is difficult to realize.
参照图1,本发明中的布斯-华莱士树型乘法器结构包括:部分积产生模块、部分积压缩模块和最终加法器模块。其中:Referring to Fig. 1, the Booth-Wallace tree multiplier structure in the present invention includes: a partial product generation module, a partial product compression module and a final adder module. in:
部分积产生模块,通过对乘数重新编码以产生相应倍数的部分积中间态、部分积符号和部分积结果。它由5个部分积产生单元021和1个改进的部分积产生单元022构成。部分积产生单元021的电路结构与传统乘法器中部分积产生单元021电路相同。该改进的部分积产生单元022电路结构如图2所示。它包括:布斯编码电路、部分积中间态产生电路和部分积结果产生电路。布斯编码电路用于对乘数重新编码以产生部分积的倍数控制信号和符号控制信号;部分积中间态产生电路根据布斯编码所产生的控制信号产生相应倍数的部分积中间态;部分积结果产生电路,用于产生部分积结果。其中部分积结果产生电路由负部分积运算电路和二选一选择器构成,它的工作原理是:负部分积运算电路对部分积中间态进行运算,产生负部分积输出到二选一选择器中,二选一选择器根据布斯编码电路所产生的部分积符号控制信号,对负部分积运算电路结果和部分积中间态产生电路结果进行选择,产生相应倍数及类型的部分积。部分积产生单元022的工作原理是:布斯编码电路产生的部分积倍数控制信号输出到部分积中间态产生电路中,部分积符号控制信号输出到部分积结果产生电路中。部分积中间态产生电路产生的部分积中间态结果输出到部分积结果产生电路中。部分积结果产生电路将所产生的正、负部分积输出到部分积压缩模块中。The partial product generation module generates the partial product intermediate state, partial product sign and partial product result of corresponding multiples by recoding the multiplier. It consists of five partial
部分积压缩模块,依据部分积压缩器阵列结构,对所有的部分积进行压缩处理,直至产生最后两行部分积,并将最后两行部分积输出到最终加法器模块中。本发明乘法器中,由于改变了部分积产生模块中最后一个部分积产生单元所产生的部分积类型,使得最后一行部分积不会将符号位带入部分积阵列中,所以与传统布斯-华莱士树型乘法器的部分积阵列相比,其部分积阵列能够减少一行部分积,从而减少部分积压缩级数,规整化了部分积阵列结构,如图3所示。在使用图3所示部分积阵列进行压缩时,压缩器阵列结构将由传统的部分积压缩器阵列结构图12改变成本发明中的图4所示。图4所示的部分积压缩器阵列与图12所示的传统的部分积压缩器阵列相比,减少了部分积压缩器阵列中压缩阵列的级数及压缩器的数量,从而简化了电路实现时的级数及电路连线的复杂度。The partial product compression module compresses all the partial products according to the partial product compressor array structure until the last two rows of partial products are generated, and outputs the last two rows of partial products to the final adder module. In the multiplier of the present invention, due to changing the partial product type produced by the last partial product generating unit in the partial product generating module, the partial product of the last row will not bring the sign bit into the partial product array, so it is different from the traditional Booth- Compared with the partial product array of the Wallace tree multiplier, its partial product array can reduce a row of partial products, thereby reducing the number of partial product compression stages, and regularizing the partial product array structure, as shown in Figure 3. When the partial product array shown in FIG. 3 is used for compression, the compressor array structure will be changed from the traditional partial product compressor array structure shown in FIG. 12 to that shown in FIG. 4 of the present invention. Compared with the traditional partial product compressor array shown in Figure 12, the partial product compressor array shown in Figure 4 reduces the number of stages of the compression array and the number of compressors in the partial product compressor array, thereby simplifying the circuit implementation The number of stages and the complexity of circuit connections.
最终加法器模块,将部分积压缩阵列所产生的最后两行部分积进行相加总得到乘积结果。The final adder module adds up the partial products of the last two rows generated by the partial product compressed array to obtain the product result.
本发明与现有技术相比,其优势在于降低了乘法器电路结构和版图实现的复杂度,同时在不增加面积的情况下,能够提升乘法器的运行速度。Compared with the prior art, the present invention has the advantages of reducing the complexity of the circuit structure and layout realization of the multiplier, and at the same time, can increase the operating speed of the multiplier without increasing the area.
Claims (4)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010013538A CN101739231A (en) | 2010-01-04 | 2010-01-04 | Booth-Wallace tree multiplier |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010013538A CN101739231A (en) | 2010-01-04 | 2010-01-04 | Booth-Wallace tree multiplier |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101739231A true CN101739231A (en) | 2010-06-16 |
Family
ID=42462759
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201010013538A Pending CN101739231A (en) | 2010-01-04 | 2010-01-04 | Booth-Wallace tree multiplier |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101739231A (en) |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105183425A (en) * | 2015-08-21 | 2015-12-23 | 电子科技大学 | Fixed-bit-width multiplier with high accuracy and low complexity properties |
CN106019129A (en) * | 2016-05-22 | 2016-10-12 | 复旦大学 | Test method for multiplier of embedded DSP in FPGA |
CN108364065A (en) * | 2018-01-19 | 2018-08-03 | 上海兆芯集成电路有限公司 | Adopt the microprocessor of booth multiplication |
CN109753268A (en) * | 2017-11-08 | 2019-05-14 | 北京思朗科技有限责任公司 | More granularity concurrent operation multipliers |
CN110378477A (en) * | 2019-08-30 | 2019-10-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110378478A (en) * | 2019-08-30 | 2019-10-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110515590A (en) * | 2019-08-30 | 2019-11-29 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110531954A (en) * | 2019-08-30 | 2019-12-03 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110554854A (en) * | 2019-09-24 | 2019-12-10 | 上海寒武纪信息科技有限公司 | Data processor, method, chip and electronic equipment |
CN110673823A (en) * | 2019-09-30 | 2020-01-10 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method and chip |
CN111025133A (en) * | 2019-10-24 | 2020-04-17 | 北京时代民芯科技有限公司 | Test method of second-order Booth coding Wallace tree multiplier circuit |
CN111258543A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258545A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258541A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258542A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258540A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258539A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN113031913A (en) * | 2019-12-24 | 2021-06-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, device and chip |
CN113031915A (en) * | 2019-12-24 | 2021-06-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, device and chip |
CN113220267A (en) * | 2021-05-20 | 2021-08-06 | 西安电子科技大学 | Booth coding bit expansion-based multiplier and implementation method |
CN114840170A (en) * | 2022-04-14 | 2022-08-02 | 苏州大学 | 2bit and 4bit Wallace Tree Multiplier Circuit Based on Resistive Memory |
-
2010
- 2010-01-04 CN CN201010013538A patent/CN101739231A/en active Pending
Cited By (35)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105183425B (en) * | 2015-08-21 | 2017-09-01 | 电子科技大学 | A Fixed Bit Width Multiplier with High Accuracy and Low Complexity |
CN105183425A (en) * | 2015-08-21 | 2015-12-23 | 电子科技大学 | Fixed-bit-width multiplier with high accuracy and low complexity properties |
CN106019129B (en) * | 2016-05-22 | 2018-11-13 | 复旦大学 | The test method of DSP embedded inner multiplication device in FPGA |
CN106019129A (en) * | 2016-05-22 | 2016-10-12 | 复旦大学 | Test method for multiplier of embedded DSP in FPGA |
CN109753268A (en) * | 2017-11-08 | 2019-05-14 | 北京思朗科技有限责任公司 | More granularity concurrent operation multipliers |
CN109753268B (en) * | 2017-11-08 | 2021-02-02 | 北京思朗科技有限责任公司 | Multi-granularity parallel operation multiplier |
CN108364065B (en) * | 2018-01-19 | 2020-09-11 | 上海兆芯集成电路有限公司 | Microprocessors for Tepbus Multiplication |
CN108364065A (en) * | 2018-01-19 | 2018-08-03 | 上海兆芯集成电路有限公司 | Adopt the microprocessor of booth multiplication |
CN111258540B (en) * | 2018-11-30 | 2023-01-20 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258542B (en) * | 2018-11-30 | 2022-06-17 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258541B (en) * | 2018-11-30 | 2022-06-17 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258545B (en) * | 2018-11-30 | 2022-08-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258539A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258543A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258545A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258541A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258542A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN111258540A (en) * | 2018-11-30 | 2020-06-09 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110515590A (en) * | 2019-08-30 | 2019-11-29 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110378477B (en) * | 2019-08-30 | 2023-09-08 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110378478B (en) * | 2019-08-30 | 2023-09-08 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110378477A (en) * | 2019-08-30 | 2019-10-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110378478A (en) * | 2019-08-30 | 2019-10-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110531954A (en) * | 2019-08-30 | 2019-12-03 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, chip and electronic equipment |
CN110554854A (en) * | 2019-09-24 | 2019-12-10 | 上海寒武纪信息科技有限公司 | Data processor, method, chip and electronic equipment |
CN110554854B (en) * | 2019-09-24 | 2024-05-03 | 上海寒武纪信息科技有限公司 | Data processor, method, chip and electronic equipment |
CN110673823A (en) * | 2019-09-30 | 2020-01-10 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method and chip |
CN111025133B (en) * | 2019-10-24 | 2022-02-22 | 北京时代民芯科技有限公司 | Test method of second-order Booth coding Wallace tree multiplier circuit |
CN111025133A (en) * | 2019-10-24 | 2020-04-17 | 北京时代民芯科技有限公司 | Test method of second-order Booth coding Wallace tree multiplier circuit |
CN113031915A (en) * | 2019-12-24 | 2021-06-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, device and chip |
CN113031913A (en) * | 2019-12-24 | 2021-06-25 | 上海寒武纪信息科技有限公司 | Multiplier, data processing method, device and chip |
CN113220267B (en) * | 2021-05-20 | 2022-04-19 | 西安电子科技大学 | A Multiplier Based on Booth Coding Bit Expansion and Its Implementation |
CN113220267A (en) * | 2021-05-20 | 2021-08-06 | 西安电子科技大学 | Booth coding bit expansion-based multiplier and implementation method |
CN114840170A (en) * | 2022-04-14 | 2022-08-02 | 苏州大学 | 2bit and 4bit Wallace Tree Multiplier Circuit Based on Resistive Memory |
CN114840170B (en) * | 2022-04-14 | 2024-07-02 | 苏州大学 | 2Bit and 4bit Wallace tree multiplier circuit based on resistive memory |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101739231A (en) | Booth-Wallace tree multiplier | |
Rao et al. | A high speed and area efficient Booth recoded Wallace tree multiplier for fast arithmetic circuits | |
Jaiswal et al. | Low power wallace tree multiplier using modified full adder | |
CN112540743B (en) | Reconfigurable processor-oriented signed multiply accumulator and method | |
Lamba et al. | A review paper on different multipliers based on their different performance parameters | |
Srikanth et al. | Low power array multiplier using modified full adder | |
CN116450217A (en) | Multifunctional fixed-point multiplication and multiply-accumulate operation device and method | |
CN103412737B (en) | Realize the gate circuit of base 4-Booth coded method and streamline large number multiplication device based on the method | |
Thomas | Design and simulation of radix-8 booth encoder multiplier for signed and unsigned numbers | |
Kakde et al. | Design of area and power aware reduced Complexity Wallace Tree multiplier | |
CN110825346B (en) | An unsigned approximate multiplier with low logic complexity | |
Sharma et al. | Modified booth multiplier using wallace structure and efficient carry select adder | |
Ismail et al. | High performance complex number multiplier using booth-wallace algorithm | |
Pathak et al. | Low power Dadda multiplier using approximate almost full adder and Majority logic based adder compressors | |
Daud et al. | Hybrid modified booth encoded algorithm-carry save adder fast multiplier | |
Laxman et al. | FPGA implementation of different multiplier architectures | |
Khan et al. | Performance analysis of reduced complexity Wallace multiplier using energy efficient CMOS full adder | |
Hussain et al. | Performance comparison of Wallace multiplier architectures | |
Murugeswari et al. | An area efficient and low power multiplier using modified carry save adder for parallel multipliers | |
Selvakumari et al. | Transistor level implementation of a 8 bit multiplier using vedic mathematics in 180nm technology | |
CN102270110A (en) | Improved 16Booth-based coder | |
Kaur et al. | Area-efficient low PDP 8-bit vedic multiplier design using compressors | |
CN111897513A (en) | Multiplier based on reverse polarity technology and code generation method thereof | |
JP4290203B2 (en) | Reduction array apparatus and method | |
Kaur et al. | Implementation of modified booth multiplier using pipeline technique on FPGA |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20100616 |